Brain structure, communication and cognition

1
Brain structure, communication and cognition

• MSc ACSB module 2005/06
• Session 8

2
Side bias in baboon threats

• Casperd Dunbar, 1996, Behav. Proc. 37, 57-65
• Gelada baboons use LVF gt RVF during fights,
  threats, approaches this leftward-bias enhanced
  if the level of aggression is high rather than
  low
• Suggests R hemisphere processing of emotional
  signals that are critical to success in these
  contests

3
Right-hemisphere and emotion?

• Hauser (1993) Science 261, 475-477
• Rhesus monkey 4 facial expressions
• Open mouth threat Ear flap threat Fear grimace
  Copulation grimace
• Asymmetry judged from videos
• Temporal onset - which mouth corner moves first
• Fullness of development (count skin-folds eye to
  cheekbone in lip retraction)
• Humans judge expressiveness of double-right or
  double-left face portraits

4
Hauser's Rhesus monkeys (2)

• Fear grimace
• 75 with asymmetric timing, LHS moves first
• Most more folds, higher mouth corner, on LHS
• Other expressions
• 50-75 of monkeys assymetric
• LHS bias also evident
• Chimaera pictures human judges find L-L image
  shows heightened fear
• Results suggest R. hemisphere dominant with
  respect to emotional expression in Rhesus monkey

5
Lateralised call processing

• Hauser, 1994, PNAS, 91, 3946-3948 1998, Anim.
  Behav., 56, 41-47
• Rhesus monkeys at food dispenser in wild.
• Play call from directly behind the monkey
• Film whether turns L or R ear to identify sound
• 4 calls
• 3 monkey calls aggressive, fearful, affiliation
• Bird (Ruddy Turnstone) alarm call as control

6
Results (Rhesus monkey in wild)

• Adults turn R ear (61/80 times) for monkey calls,
  L ear (13/15) for control (bird calls)
• Infants show no bias for any call
• Indicates that adults are processing own calls
  preferentially using left hemisphere, just as do
  humans with language and that restriction to L
  Hem occurs over course of development

7
Japanese Macaque smooth coos

• Smooth Coos are one of 7 types of Coo call in the
  monkeys repertoire
• Early and Late Smooth Coos differ in peak
  position

8
Japanese monkeys (1)

• Petersen, 1978, Science 202, 324-327
• Lab. training task Early and Late Smooth Coos
  played into R or L ear at random, competing
  wide-band noise played in other ear must
  discriminate for reward using either PEAK
  POSITION or PITCH
• Showed REA (L Hem processing) in detecting early
  vs. late peak position in own calls, but not for
  pitch of calls
• Other monkey species showed no REA for these calls

9
Japanese moneys (2)

• Lab learning data suggest Japanese monkeys
  process coo calls in L Hem
• Ablation of auditory cortex on L impairs
  discrimination, on R leaves it unaffected
  (Heffner)
• If L cortex removed, can re-learn but if both
  sides removed, cannot re-learn

10
Implications

• Lateralisation of processing not unique to human
  language may be more widely distributed for
  efficient processing of species own
  communication signals
• What about processing of over-learned signals of
  another species, e.g., where apes/monkeys/dogs
  understand some human speech?

11
Categorical perception ?

• Humans show categorical perception of syllables
  e.g. /da/ vs. /ta/ (differ in VOT)
• Jap. monkeys trained on early vs late smooth
  coos,
• E1 vs L1, then add additional Es or Ls so that
• (E1, E2) vs L1, or E1 vs (L1, L2) until reach
• (E1..E8) vs (L1..L8) eight of each in set
• Test on intermediate calls (synthesised)

12
Categorical perception (2)

• Show categorical perception (rapid change in
  responses) for peak position

13
Categorical perception (3)

• Jap. monkeys find peak position easy to learn as
  set size increased, but find pitch hard to learn
  (not naturally relevant)
• Other species find pitch of Jap. Monkey coos
  easier to learn than peak position

14
Neural control of birdsong

• Birdsong controlled by brain nuclei and motor
  commands down XII (hypoglossal) nerve
• Nottebohm cut XII nerve unilaterally
• L abolishes all major components of song
• R only minor losses

15
Song control (2)

• HVC (higher vocal centre is major control centre
  for song)
• L HVC lesions have major impact, R HVC do not.
• Left lateral control of the major/ complex sounds
  in the song many species, but not all zebra
  finch has R sided control
• Could the L brain be specialised for control of
  complex motor patterns?

16
Brain size - ability relationships

• Bigger animals have bigger brains (elephant vs.
  mouse)
• Relate brain size to body size (Jerison)
• When body size accounted for, differences in
  brain size can be related to
• Diet
• Evolutionary relationships
• Complexity of social (group) environment

17
Brain areas and abilities

• Specialised brain areas may expand to enhance
  cognition/ communication
• Szekely warblers with bigger song repertoires
  have larger song control area in brain to manage
  the bigger repertoire
• Lefebvre feeding innovation and forebrain size
• Sherry Gaulin larger hippocampus in
  food-hoarding birds and polygamous Microtus spp.

18
European warbler song repertoires

• Székely et al., 1996, PRSB 263, 607-610
• Correlation r0.68 between syllable repertoire
  size and HVC residual volume
• Used phylogenetically independent contrasts

19
Feeding innovations and forebrain volume

• Lefebvre et al., 1997, Anim. Behav. 53, 549-560.
• Bigger forebrains in birds that are more often
  reported to use innovative feeding techniques

20
Hippocampus in food-hoarding birds
Spp. of bird that make and recover food-caches
need a better memory for location than spp. that
find food naturally. Food-hoarding spp. bigger
AVIAN HIPPO-CAMPUS than expected for body weight
(Sherry)
21
Tactical deception and neocortex ratio
22
Neocortex and mating success in NHPs

• In dull polygamous NHPs, rank (fighting ability,
  size) will determine male mating success
  dominants win and mate
• But in brighter species, non-dominants may make
  alliances and so improve their mating success
• Pawlowski et al (1998) Behaviour 135, 357-368 a
  negative correlation between neocortex size and
  mating success
• Males of spp. with large neocortex exploit
  opportunities to undermine the dominant's
  power-based monopolisation of ovulating females,
  unlike spp. with smaller neocortex. This effect
  is independent of male cohort size.

23
Implications of brain-size data

• Greater/specialised capacities depend on greater
  brain size overall, or in relevant specialised
  areas
• Brain is not purely a general-purpose processing
  device
• May have specialisation for communication or
  social problem solving

24
Dunbar brain size, group size, gossip

• Dunbar uses NHP brain size-group size
  relationship to predict human group size (150)
  from human brain size
• Argues that grooming required to maintain social
  bonds in group of 150 would take too much time
• Conversation group allows bonding with several
  others at once, cuts time-cost of keeping up the
  bonds gossip, like grooming, is reinforcing and
  enjoyable. Focus not on unknown information

25
References

• Hauser (1997) The evolution of communication. Ch
  4 pp. 175-211 Ch 7 pp. 534-548.
• Petersen (1982) in Snowdon et al. (Eds) Primate
  communication. Ch. 8
• Barton (1997) in Whiten Byrne Machiavellian
  intelligence II. Ch. 5
• Casperd Dunbar (1996) Behav. Proc., 37, 57-65
• Byrne Corp (2004) Proc. Roy. Soc. Lond., B,
  271, 1693-1699
• Garamszegi Eens (2004) Ecology Letters, 7,
  1216-1224 Healy, de Kort Clayton (2005) TREE,
  20, 17-22