Home | JumpStation | General | Research | Undergrad | BBS | Postgrad | IT| Philosophy FAQ | On-Line Phil Papers

 
INTRODUCTION TO COGNITIVE SCIENCE
NEUROSCIENCE COMPONENT

REPRESENTATIONS IN THE NERVOUS SYSTEM

IC Bruce (hrmybic@hkucc.hku.hk)

 

LECTURE 1: Real neurones: How do they work?

The brain is an electrochemical machine. The components of the machine – the neurones - are relatively simple; the ways in which they are interconnected produce extraordinary complexity. Information is represented in the brain, & moved around the brain, in the form of brief electrical pulses – action potentials. Neurones communicate with each other by releasing chemicals – neurotransmitters – across small gaps between neurones - synapses. Neurotransmitters are of two kinds: excitatory neurotransmitters increase the probability that a target neurone will generate an action potential; inhibitory neurotransmitters decrease the probability that an action potential will occur. At any given moment, the sum of the inputs to a neurone determines whether or not it generates an action potential. Each action potential lasts about 1 millisecond & so a neurone can operate at a theoretical maximum frequency of 1,000 Hertz (in reality, few neurones ever exceed 500 Hertz); action potentials are conducted along processes of the neurones – axons – at velocities up to 120 meters per second. There are said to be about 1014 neurones in the human brain, each receiving about 3,000 inputs & projecting to about 3,000 target neurones.

 

LECTURE 2: Sensory receptors: How do they encode information about the environment?

Sensory receptors are transducers. They convert different forms of energy in the environment (energy in the forms of light, chemicals, mechanical force & temperature) into neural codes (sequences of action potentials in specific pathways). Our perceptions are limited by the properties of our receptors. For example, our light receptors – rods & cones in the retina – are only sensitive to a small range of frequencies in the electromagnetic spectrum that we call "visible light". We cannot sense higher (e.g. X-rays) or lower (e.g. radio waves) frequencies. Information about the environment is encoded in three ways: (i) as labelled lines – action potentials in the auditory pathway are perceived in terms of sound & action potentials in the visual pathway are perceived as light. This is a consequence of anatomical organisation; (ii) as population codes – weak stimuli activate a few receptors, strong stimuli activate many receptors; & (iii) as a frequency code – for any single receptor, a weak stimulus generates a few action potentials while a strong one causes many action potentials.

 

LECTURE 3: The Brain: Our Universe Within. #1. Evolution. (Video)

 

LECTURE 4: Maps in the cerebral cortex: How are they organised?

The cerebral cortex forms a convoluted layer 2-5 mm thick, with a total surface area of about ¼ square meter - that is, about the same area as a table for two at Macdonald’s. Each cubic millimetre of cortex contains about 90,000 neurones, 400 meters of dendrite (the input surface of the neurone), 3.4 kilometres of axon, & 700,000,000 synapses.

Maps are not accurate copies of places. A good map indicates the locations of significant items (e.g. streets, not trees). Significant items are filtered out from the constant stream of sensory input arriving from the receptors. The cerebral cortex operates in a modular format, each module performing specific operations on its inputs & sending the results to the appropriate target modules. There are maps of the retinas in the occipital cortex (at the back of the brain), of the cochleae (the receptors for hearing) in the temporal cortex (at the sides of the brain), & of the skin, muscles & joints in the parietal cortex (at the top of the brain). These maps process incoming information & send the results to more abstract maps. In the association cortex, information converges from the various senses to create maps that represent such things as body image. The frontal cortex includes maps related to voluntary movement (including speech) as well as association areas involved in planning.

 

LECTURE 5: Maps in the cerebral cortex: How does cognition appear?

At any given moment, the combination of serial (hierarchical) & massively parallel processing means that many modules in the cerebral cortex are active. At some point, this mass of activity produces what we would identify as a cognitive event, such as recognising your old friend’s face or deciding to run for the minibus. Current experimental studies focus on revealing some form of synchronisation between active modules underlying this phenomenon. Synchronisation at 40 Hertz is the theme of the moment.

 

LECTURE 6: The Brain: Our Universe Within. #2. Perception. (Video)

 

TUTORIAL: Simple neural networks (CAL); Shadow speech experiment.

 

ASSIGNMENT: Write no more than 500 words on the key concepts related to either "Functional asymmetry: Left brain – right brain" or "Cortical representation of speech". Submit your answers on or before Friday 20 November, 1998.

 

REFERENCES: