The Big Question

1
The Big Question

• How is the behavior known as auditory imagery or
  imagination produced by the brain?

2
Theoretical Questions

• Is there an imagination module or network?
• Would you expect partly or wholly different
  results working in different modalities?

3
Theoretical Questions

• Should such areas be similar to or include
  perceptual areas in the modality being tested?
• Conceptual support sharing similar hardware
• Empirical support Zattore, Farah, Kosslyn, Chen,
  
• In 1998 Chen et al. Found that LGN could be
  activated by visual imagery

4
Theoretical QuestionsTowards a Hypothesis

• Are there any areas that are more interested in
  imagination than real perception
• Are there any areas that dont care about the
  modality of the imagined stimulus, but seem to be
  involved in any kind of imagery

5
Hypothesis

• We would find one or more areas which responded
  to both real and imagined stimuli, as well as
  areas responding uniquely to imagination.

6
Auditory Background
Superior Temporal Gyrus Zatorre 1994
RW-14
7
Auditory Background
Inferior Frontal Gyrus Zatorre 1996
Midfrontal Area (Precentral Gyrus) Zatorre 1996
RW-18
8
Auditory Background
Right Occipital Cortex (Fusiform Gyrus) Zatorre
1994
RW-05
9
Imagination Background
Precentral Gyrus Halpern 1999
RW-18
10
Imagination Background
Supplementary Motor Area Halpern 1999 Rao 1997
RW-22
11
Imagination Background
Left Frontal Pole Halpern 1999
L. Middle Frontal Gyrus Halpern 1999
L. Inferior Frontal Gyrus Halpern 1999
R. Inferior Frontal Gyrus Zatorre 1994
RW-10
12
Imagination Background
Medial Anterior Cingulate Gyrus Zatorre 1994
RW-18
Right Anterior Cingulate Gyrus Zatorre 1994
RW-15
13
Imagination Background
Heschyls Gyrus (Primary Auditory Cortex) Decrease
d Activity Halpern 1999
Associative Auditory Cortex Halpern 1999
RW-13
14
Experimental Design6 Seconds of Pure Tone
6
15
Experimental Design6 Two-Second Math Problems
18 - 9
6
12
16
Experimental Design6 Seconds of Imagining Pure
Tone
6
12
6
17
Experimental Design6 Two-Second Math Problems
14 - 8
6
12
6
12
18
Theoretical Design Issues

• Just what is our task?
• Imagery
• Auditory Memory / Retrieval
• Working Memory Shifts / Attention?
• Timing
• Pitch Discrimination?

19
Analysis MethodsCorrelation Analysis

• Events are spaced wide enough for a hemodynamic
  response to be observed
• Data has significant amount of high-frequency
  noise

20
Analysis MethodsDiscrete Fourier Analysis

• Events occur periodically
• Strong high-frequency noise makes other analysis
  methods difficult
• Used to interrogate an established auditory ROI
• The data often have a strong first harmonic
  seemingly unrelated to the experiment

21
Analysis MethodsEvent Related Potentials

• Averaging produces clean hemodynamic curves in
  presence of strong noise
• Relative activation strengths can be compared
• Single instances of very strong noise (like head
  movement) result in skewed curves

22
Precentral Gyrus ActivationCorrelation Analysis
RW-18
RW-18
Imagination
Auditory
23
Superior Temporal GyrusCorrelation Analysis
RW-12
RW-12
Imagination
Auditory
24
Precentral Gyrus ActivationCorrelation Analysis
of Auditory Localization Run
RW-18
PM-18
25
Precentral Gyrus ActivationCorrelation Analysis
of Auditory Localization Run
BH-13
TW-16
26
Precentral Gyrus ActivationFourier Evidence (PM)
27
Precentral Gyrus ActivationFourier Evidence (BH)
28
Regional ActivationEvent Related Potentials
Precentral Gyrus
Auditory Cortex
29
Regional ActivationEvent Related Potentials
Real Sound
Precentral Gyrus
Auditory Cortex
30
Regional ActivationEvent Related Potentials
Imagined Sound
Precentral Gyrus
Auditory Cortex
31
Regional ActivationEvent Related Potentials
Precentral Gyrus
32
SMA ActivationEvent Related Potentials
33
SMA ActivationEvent Related Potentials
34
Other Areas of Activation
RW-07
35
Putamen Activation
RW-10
36
Putamen Activation
RW-11
37
Putamen Activation
RW-12
38
Supplementary Motor Area Activation
RW-22
TW-19
39
Integrating Ideas

• Other tasks involving these areas and behaviors
• i.e.Mental Timing Rao et al. (1997)
• (L) Listen passively
• (S) Tap finger in synch w/ music
• (C) Continue tapping at same tempo but w/o music
• (D) Discriminate between pitches
• Results
• S and C Motor cortex (PcG), r. cerebellum, STG
• C only SMA, Putamen, left vl. thalamus, r. IFG
• C and D STG (caudal)
• D SMA (rostral)

40
Integrating Ideas

• Reinterpreting tasks
• Note auditory cortex activity under continuation
  condition - does this imply auditory imagery is
  being used, as in our task
• Auditory imagery requires timing - could our
  results be due to a mechanism for timing and not
  imagery per se
• Although the continuation condition did not
  involve an auditory stimulus, internal rehersal
  of the tone interval duration, or auditory
  imagery, is likely to be used during this
  performance condition Rao et al.

41
Integrating IdeasFinding a Neural Model for
Auditory Imagery

• Subdividing the Mechanism
• There may be areas specific to each part of
  auditory imagery, i.e. right STG and right IFG
  involved in retrieval and rehearsal, rostral SMA
  involved in pitch discrimination

42
Integrating IdeasFinding a Neural Model for
Auditory Imagery

• Articulatory Loop, Working Memory Ideas
• Rehearsal of letter strings activates IFG and STG
  also (Paulesu et al., 1993)
• Is this the effect of rehearsal or imagery
  involved in rehearsal?
• If articulatory loop, would you expect areas to
  be active continuously in silent interval
• Does not fit our finndings
• Would we have found such areas if they do exist?

43
SMA ActivationEvent Related Potentials
44
SMA ActivationEvent Related Potentials
45
Integrating IdeasFinding a Neural Model for
Auditory Imagery

• Articulatory Loop, Working Memory Ideas
• Did math problems prevent articulatory loop?

46
Integrating IdeasFinding a Neural Model for
Auditory Imagery

• Motor activation ideas
• Note preference for traditional motor related
  areas
• Precentral gyrus
• Putamen
• Supplementary Motor Area
• Cerebellum

47
Integrating IdeasFinding a Neural Model for
Auditory Imagery

• Similarities that might be meaningful
• Crawford et al. (1999)
• Showed SMA, precentral gyrus, and putamen
  activation in an imagination / attentional /
  disattentional task?
• Even in so different a task, it is these
  motor-related areas which get involved in
  imagery.
• Could this represent a universal network for
  imagination?
• Might it have something to do with motor activity?

48
Integrating IdeasFinding a Neural Model for
Auditory Imagery

• So maybe
• Huge Speculation Could motor plannin be the
  basic mechanism for imagination
• In terms of motor planning, people are acting as
  though they are doing / producing the imagined
  task, but the actual motor movement is inhibited
  somewhere (Cerebellum?)

49
Future Research

• More deliniating timing, attention, working
  memory, imager, articulation

50
Summary and Conclusion

• It seems that there are areas which respond to
  imagination as much or more than to a real
  stimulus, supporting our hypothesis
• Auditory imagery also activated known auditory
  areas (STG) as well
• The mechanism, regardless of its nature, seems to
  involve the SMA, Precentral Gyrus, Putamen, and
  STG

51
(No Transcript)