Question 6; Can machines have conscious minds?

25 important questions on Question 6; Can machines have conscious minds?

Which argument is fatal for the identity theory?

The argument op multiple realizability, which is the idea that mental states can be constructe in different ways and made of different stuff.

What is consciousness chauvinism?

Only creatures that have human brains can have a conscious mind.

What does it mean that mental states are multiply realizable with regards to pain?

That there are different ways to realize pain. W can make things from different materials in different ways.
  • Higher grades + faster learning
  • Never study anything twice
  • 100% sure, 100% understanding
Discover Study Smart

What does it mean when the ''is'' of identity is symmentrical?

If water is H2O, then H2O is water.

What do functionalists claim?

THat mental states are constituted by their causal (or functional) relations to sensory input, other mental states and behavioral output.

What is folk psychology?

The common sense psychology that we use in everyday life to explain and predict the behavior of other people.

What are 3 properties of a propositional attitude (PA)?

  • They are about something.
  • Each PA is a mental state on its own.
  • They are functionally discrete: we can learn a new propositional attitude to forget one without altering any of the other propositional attitude  .

What are the elements of a Turing Machine?

  • A tape that fuctions as the memory of the machine.
    • Different sections each contain one symbol.
  • Machine can read, erase or overwrite the symbols.
  • What the machine does depends on its internal state, the input and the machine table: a set of rules that determine what will happen if the machine is in a certain state and reads a certain symbol. That is the machine's programme.

When does a computer pass the Turing test?

When it succeeds in fooling a human.

What does it mean when a computed passed the Turing test according to Turing?

That the computer is capable of thinking.

What is the problem with Shakey's computer?

It took a really long time to calculate and come up with the map and the plan which is not biologically realistic.

Whatis the problems with machine functionalism?

It is not a biologically realistic model of thinking since it processes information in a serial fashion it causes the machine to have a low damage tolerance.

What are two problems with the Turing test?

  • Syntax does not lead to semantics (with the appropriate set of rules to manipulate symbols the computer will not be able to actually understand anything).
  • Animals cannot read and write and therefore they will not be able ot pass the test. 

What does place constraints on the supervenience base?

That some types of material cannot realize rigidity.

What does it mean if a theory is brain-centric?

That the mental ''software'' runs on the ''hardware''.

What is the problem with brain-centric theories?

Maybe other parts of the body also belong to the supervenience base of the conscious mind. If that indeed is the case, then it is not sufficient to look at only the brain in order to understand the mind.

What are the problems with functionalism?

  • It is biologically unrealistic.
  • The model is absed on syntax, but that does not lead to semantics.
  • The argument based on multiple realizability might not be as strong as some think.

What are the two types of connectionalists?

  • Scientists who actually build brain-like netwokrs.
  • Philosophers wo argue that these connectionist models provide insight into how the mind works.

Name and explain all the parts of neural networks.

  • Neurons: send information to other nerons and other types of cells through electrical pulses and neurotransmitters.
  • Soma: the cell body of a neuron consisting of two extensions.
    • Dendrites: shaped like a tree and receive information from other neurons in the synapses.
    • Axon: place where the signal enters even more synapses and the signal is transmitted via neurotransmitters like dopamine and serotin to other neurons.

What is backpropagation / gradient descent learning?

When knowing what the output should be, it is possible to adjust a weight and see if the output vector changes the right direction. If this is the case, one can change a weight again, and see if the result is better again. If it gets worse, then chang it back to the latest settings and make som other random change.

What is snapshot reasoning?

It judges whether this one input is a face and gives just one output but it does not deal with any changes in time.

How can we solve the snapshot reasoning problem?

Add an extra layer: a recurrent network. If information is then sent form the hidden layers to the output units and some other units called context units, these context units feed the information back into the hidden layer a moment later.

How are things represented in a connectionist network?

Representations are distributed over the system, not that each time a representation is added to the memory of the system that it takes up a part of the memory.

What sets off the creation of an artificial system that can actually understand the input?

The syntax and semantics are closely related to each other and the input.

What are the problems for connectionism?

  • Thought is systematic, which means that it is rule-based, just like language, and uses discrete concepts or words.
  • It is a brain-centric view and although it is not clear that this view is wrong, but it is not evident that it is true either. 

The question on the page originate from the summary of the following study material:

  • A unique study and practice tool
  • Never study anything twice again
  • Get the grades you hope for
  • 100% sure, 100% understanding
Remember faster, study better. Scientifically proven.
Trustpilot Logo