Chess experts require less brain activity than nonexperts in a recognition task with chess stimuli

1
Chess experts require less brain activity than
non-experts in a recognition task with chess
stimuli 

• Guillermo Campitelli,

Fernand Gobet, Amanda Parker School of
Psychology, University of Nottingham
2
Main results in chess research

• Long term memory storage of chess patterns
• (Chase Simon, 1973 Gobet Simon, 1996).
• Reconstruction experiments superiority
• experts vs novices.(e.g., deGroot 1946)
• Automatic access to long term storage.
• (CHREST, Gobet, 1998)
• Use of this long-term storage alleviates
• working memory demands.(Saariluoma, 1991)

3
Global workspace

• Baars (1983).
• Conscious perception involves more than sensory
  analysis it enables access to widespread brain
  sources, whereas unconscious input processing is
  limited to sensory regions.
• Baars (2002)
• Practice reduces metabolic activity.

4
WM/LTM in the brain

• Working memory processing
• frontal and parietal areas (e.g. Duncan Owen,
  2000 Goldman-Rakic, 1994 Fuster, 1998)
• Long-term memory recognition
• inferior temporal areas (e.g. Logothetis, 2002
  Tanaka Taylor, 1991).

5
fMRI study

• 1 group of chess players (5 subjects)
• 2 Masters, 1 Expert, 2 Class B players.
• 1 group of non players (7 subjects)
• Conditions
• Game
• Dot
• Scene
• Random

6
Block structure
12.5 secs.
2 s.
6.5 secs.
3 secs.
1 s
3 secs.
7 test stimuli
7
GAME
DOT
SCENE
RANDOM
8
Processing

• 3T scanner (Magnetic Resonance Centre,
  University of Nottingham)
• T2 weighted Echo-Planar images (EPIs)
• TR 3 seconds
• 22 coronal slices at 136ms each.
• 756 volumes acquired (first 4 discarded)
• 44 blocks (11 per condition)
• Voxel size at acquisition 3 x 3 x 9
• Normalised to 3 x 3 x 3 (SPM 99)
• Smoothing 8 x 8 x 8 (SPM99)
• Box-car blocked design

9
Hypotheses

• Game gt Scene
• Chess players activation of long-term
• memory storage areas (inferior and
• medial temporal lobe)
• Game gt Dot
• Chess players low activity in working
• memory areas (parietal and frontal
• areas).
• Non-players high activity in working
• memory areas (parietal and frontal areas).

10
Results. Accuracy
11
Results. Response times
12
Results. Game gt Scene
CHESS PLAYERS
NON - CHESS PLAYERS
238 vs 0 voxels
13
Results. Game gt Dot
NON - CHESS PLAYERS
CHESS PLAYERS
392 vs 4700 voxels
232 vs 4010 voxels
14
Results. Game gt Random

• Only 16 voxels activated in left parahippocampal
  gyrus in chess players.
• A pattern of right frontal and anterior temporal
  areas in the non-chess players.

15
Discussion

• Beck et al. (2001).
• Awareness requires fusiform and parahippocampal
  gyri (ventral pathway) and parieto-frontal
  activation (dorsal pathway)change detection
• Activation of the ventral pathway alone does not
  reach consciousness. change blindness
• Jansma et al. (2001).
• Practice reduces frontal and parietal activity.
• Weissman (2002).
• Practice reduces activity in fronto-parietal
  areas, but if the previous practice leads to
  conflict (automatic schemas are triggered when
  they are irrelevant) frontal activation
  increases.

16
Discussion (cont)

• Gauthier et al. (1999). Activation of fusiform
  gyrus increases with expertise recognizing novel
  objects.
• Gauthier et al.(2000). Activation of fusiform
  gyrus increases in experts in birds and cars when
  recognizing domain-of-expertise stimuli.

17
Conclusion

• Increased activity in inferior and medial
  temporal areas (game gt scene) in chess players
  activation of LTM patterns related to the field
  of expertise.
• Decreased activity in frontal and parietal areas
  (game gt dot) in chess players Automaticity makes
  the experts to require less conscious control to
  perform the task.

18
Conclusion

• Higher conscious control requires higher brain
  activity in frontal and parietal regions.

19
END
20
Results. Game gt Scene
21
Results. Game gt Dot
22
Results. Dot gt Game