07 Functional Systems

Hadrocodium wui 195 mya (~10m Neurons, close relative to mammals)

skull forms with brain

fur: sensory modality -> bigger somatosensory cortex needed

mammals: longer development in body than in egg -> longer for the brain to develop (also after birth)

biggest driver: olfactory bulb and neocortex

reptile like mammal ancestors + sister groups + mammals

synapsid brain with similar main subdivisions: olfactory bulb, cortex with lateral olfactory part, dorsal sensory part, and medial region for memory functions

• medial region in humans: hippocampus

neocortex with 1 dominating sensory field: visual

dorsal somatic input

corpus callosum (placental mammals)

motor commands (efferences) from motor and somatosensory fields

cortex piriformis becomes proportionally smaller with phylogeny

~15 brain sections

new-world monkeys: dichromatic vision, flot nose, grabbing with tail

old-world monkeys: downwards facing nostrils, 25mya -> apes

social groups -> cognitive challenges

• visual system to recognize mimic and faces

auditory and motor system

prefrontal area for planning, intentions

• humans: even bigger, relatively same size

prefrontal cortex

• more complex dendritic arborizations, more spines

  etc.

  • personality, decision making, basics of sociality, social control, orchestration of thoughts

    and actions

larger posterior parietal cortex: associative area - convergence of visual and somatosensory input -> decision making and tool use

• right PPC: dominant role in visual spatial attention, reading, calculating, coordinate

  transformation

  • hemineglect

• coordination stability (e.g., moving head do not lead to a loss of orientation in a room)

• PPC transforming representations (e.g.,

  musical key transformation)

more cortical areas (150-200)

but progressive loss of olfactory receptors

less genetic determination

left: langugage, calculation

right: prosody, music, face recognition

long transmission time between hemispheres

(feedback) loops

CDIs: corollary discharge interneurons

reflex inhibition in snail

sensory filtering in crickets

sensory analysis in macaque: attention preceeding receptice field slightly

predictions: bats comparing own output to echo

computastional theroy: what and why

representation and algorithm: how

• input -> output

hardware implementation: physical realization

decision variables with weights -> threshold = decision boundary

line of decision boundary by adding layers of neurons

• axes stretched in space

  • higehr layer: higher level of representation

network learning from the brain to decode real human decoding pattern

30x30 retina: first layer with 500 neurons

second layer: 500 neurons computing activity pattern of first layer

third layer: 2000 neurons complex statistical dependencies

actual number in output layer

reversing system: generating new numbers - stochastic system

informativeness: provide useful evidence

• programme playing, no evidence it knows the rules

trust: errors

• self-driving cars