Honest Signalling, Is Speech Production and Perception Special? Flashcards Preview

Evolution of Vocal Communication > Honest Signalling, Is Speech Production and Perception Special? > Flashcards

Flashcards in Honest Signalling, Is Speech Production and Perception Special? Deck (27):
1

In handicap signals, the signal depends on..

Johnson 97

.. a balance of costs and benefits involved for the signaller

there is more benefit to one of high need

2

Parent/offspring relationship

Furlow 97
- infant condition correlated with acoustic structure
- crying is more costly to an infant of lower phenotypic quality, parents more likely to respond to cries of lower quality

3

Types of cost? How does the second relate to handicap?
eg. Petrie
eg. animal

efficacy - cost of making signal
strategic - over and above the cost of signalling itself to make sure it ensures honesty
handicap when strategic cost is high for low quality individuals
Petrie: longer tail feathers in peacocks females laid more eggs
spotting springback - jumps away from predator to show its fit but also could just run away

4

Types of signal cost?

developmental costs - on neural circuits
production costs - effort and cost of being noticed by predators
maintenance costs - of bearing the signal eg. begging in children

5

Griffen 01

groan of pain model - vocalisations are automatic in animals

6

BUT Fitch

audience effects
eg. no alarm calls if no con-specifics

7

Cherry and Seyforth

SO..

mother monkeys don't change their behaviour when there's a predator or food that their child does or doesn't know about or

animals vocalise to act on behaviour not on minds

8

McComb et al. 94

playback experiment to generate controlled, fake contests between groups of lions
defending lions were less likely to approach calls from groups of 3 lions than 1 and approached more cautiously
decision to approach was based on the size of their group, and would attempt to recruit extra companions for approaching by roaring

9

What does categorical perception say about speech perception being special?

eg. VOT

We have sharp ID boundaries for phoneme categories for learning language -- can warp acoustic space specific to our language

shorter VOT for /p/ longer for /b/, can ID the difference around 16ms VOT

10

Kuhl 04

infants can discriminate Hindi and Salish sounds at 6 months but not 12
we have a magnet effect for native language sounds
we have computational strategies to detect statistical and prosodic patterns, and input leads to discovery of new words

11

Kuhl and Miller 78

so..?

chinchillas show categorical perception with human speech

speech adapts to an already existing system

12

What are formants?

groups of selectively enhanced harmonics, like vowels in humans

13

When humans split from chimps..

.. lost the laryngeal air sac, and so now we have descended larynxes -- increases risk of choking so should be some advantage, but not just in humans so not special to language
BUT the tongue root is retracted and rests on larynx, giving more degrees of freedom, so is probably where the ability to speak comes from

14

3 hypotheses for permanently descended larynx

mouth breathing? but other animals do this
by-product of walking upright? but kangaroos
size-exaggeration? permanent signal of larger size, may explain sexual dimorphism (m VT longer during puberty = lower pitch)

15

Ohala

size exaggeration -- purpose of enhancing threat displays

16

Fitch's hypothesis

VT formants should be honest signal acoustic cues to signallers size, as determined by length of vocal tract which is hypothesised to scale reliability to overall body size

17

SUPPORT

Plotsky et al. 13
- Portugese water dogs and Russian silver foxes
- X-rays of vocal tract correlated with body size
- suggest elements for honest and deceptive signalling if using the oral or pharyngeal components, because the pharyngeal component didn't correlate

18

SUPPORT

Rendall et al. 05
- in adult humans and baboons, males were larger than females and had lower mean voice fundamental and formant frequency, buuuut beyond this, variation didn't track body size variation
+ wide age range 18-44 years
BUT not natural speech

19

Van Dommelen 93 buuuuut

BUT Smith 03

in human men, pitch doesn't correlate with body size buuuuut they just looked at fundamental frequency not formant

BUT we still treat it as if it were a reliable cue

20

Feinberg et al. 03

women more attracted to voices with lower formant frequencies -- attraction to full sexual maturity?
but horny college students may not be representative

21

Fitch and Reby 01

red deer stags permanently descended larynx able to drop lower in breeding season, increasing vocal tract length. may come from selection favouring those able to exaggerate their size

22

Reby and McComb 03

Limitations

57 stags in adult and subadult groups, guessed weight but reproductive success known
fundamental frequency didn't give clear indication of body size but formant frequencies correlated with body weight which in turn can be used to indicate body size and vocal tract length
SO positive correlation between vocal tract length, age and weight

more adult than subadult used
only guessed body size and weight as didn't have recorded for year of study

23

Reby et al. 05

Indicates

PROS

CONS

stags were more attentive to and gave lower formant frequency roars extending their vocal tract in response to recorded roars manipulated to have lower formant frequency and thus indicate larger body size

--> role in assessment of rivals, acts as a means to assess opponent size
--> possible evolution origin for perceptual relevance of formants in humans

unfamiliar with the calls as they would be irl but also manipulated may not sound natural

24

Dunn et al. 2015

trade off between investments for reproduction
howler monkey species with harem groups have large vocal tracts and small testes
those in multimale groups have small vocal tracts and large testes -- higher levels of postcopulatory sperm competition

25

Deciet

sender benefits, reciever is harmed
when one animal exploits another to improve its fitness

26

Flower 11

the Drongo impersonates calls of other species to steal their food

27

Batesian mimicary

frog pretends to be poisonous so predator takes longer to learn which frogs it can and can't eat