Cognitive Science Exam
Which of the following are examples of cultural artifacts that allow us to practice mind-reading? a. Looking at the posture of a statue and trying to figure out what it would be thinking or feeling b. Reading a story and trying to figure out what its characters are thinking or feeling c. Reading your favorite blog because you like how its author thinks d. Watching a sporting event to figure out how teammates are sharing intentions and working together
All of the above
In a connectionist network, how are the layers connected?
Each input goes to each possible output
In your own words, explain why evolution designed us so that the SEEKING (or WANTING) system was stronger than the LIKING system. If possible, given an example of the strength of the SEEKING system from your own life. (The example can be hypothetical)
Evolutionarily, I would say the seeking or wanting system is stronger than the liking system because, we as humans, have a desire to feel needed in some way, shape, or form. Seeking or receiving exciting information makes us feel desired and like we are important, ie getting an email from a friend. In my life, something that is similar to these examples would be that I love to learn new things about topics I am unfamiliar with. That being said, I would sometimes spend a half hour going down a rabbit hole about the octopus or oceans or political topics and it seems like only 5 minutes have gone by.
T/F: The article suggests that internet surfing is addictive because it stimulates the pleasure (or LIKING) region of the brain.
False, (The article suggests it's addicting because it stimulates the SEEKING (or WANTING) region of the brain.)
T/F: Zunshine maintains that "mind-reading" always involves our trying to consciously figure out what another person is thinking or feeling.
False, (This is one of the "misconceptions" about mind-reading that she identifies.)
T/F: The only way that an AI system can solve a problem is by exhaustively tracing every branch within the search space until a solution is found.
False, Newell and Simon designed a computer (GPS) that instead relied upon "heuristic search techniques" that do NOT perform such exhaustive searches.
T/F: One advantage of symbolic AIs is their ability to learn things on their own.
False, Symbolic AIs typically need their programs to pre-program most of their knowledge!
T/F: According to the Physical Symbol System hypothesis, Artificial Intelligence is impossible
False, The hypothesis says that if a computer realized the correct system of physical symbols and rules it would, in fact, be intelligent!
T/F: If a physical symbol system is using a binary language that means that if you look inside the machine you should find literal "0"s and "1"s in its hardware.
False, What you will find instead are some kind of recurring physical patterns that correspond to the "0"s and "1"s.
physical symbol system hypothesis in action (which robot portrays the this system?)
SHAKEY- The first ever "electronic person"... Has eyes, a memory bank, feelers, and can move objects
Define weak AI
The project of constructing AI systems that can solve problems perfectly
Define strong AI
The project of constructing autonomous software that is as intelligent as human beings
connectionist network
The type of network proposed by the connectionist approach to the representation of concepts. Connectionist networks are based on neural networks, but are not necessarily identical to them. One of the key properties of a connectionist network is that a specific category is represented by activity that is distributed over many units in the network. This contrasts with semantic networks, in which specific categories are represented at individual nodes.
T/F: According to Newell and Simon, intelligence involves the ability to successfully navigate a search space.
True
T/F: Deep Blue was a symbolic AI.
True
T/F: Dopamine is a neurotransmitter that activates the SEEKING system. More dopamine = more SEEKING, less dopamine = less SEEKING.
True
T/F: Newell and Simon maintained that intelligence was a general phenomenon that could be applied across multiple domains. Because it is limited in terms of the operations it can perform, a pocket calculator does not count as intelligent.
True
T/F: Panksepp pointed out that the region of the brain stimulated (accidentally) by Olds caused rats to sniff and forage. On the basis of this, Panksepp argued that this region of the brain isn't a pleasure center. Instead, it's a SEEKING center.
True
T/F: The article suggests that the "way forward" to developing truly intelligence AI is to merge the symbolic approach and the neural network approach.
True
T/F: Zunshine argues that if we all suddenly stopped being mind-readers, much of our culture would cease to be.
True
T/F: Zunshine argues that the body is a primary source of information about the minds of other people, but the body can also be used to deceive others when they try to read our minds.
True
T/F: Zunshine argues that we are constantly trying to mind-read, but that this fact does not guarantee that our mind-reading efforts will always be accurate.
True, She gives some nice examples of how cultural differences can lead to incorrect mind-reading.
T/F: It's relatively easy for human beings to understand the reasoning process used by symbolic AI. It can be harder for human beings to understand how a neural network reaches its conclusion.
True, This is because we (human beings) program a symbolic AI to use the logical principles that it uses to reach new knowledge. In the case of neural networks that are trained by backpropagation, however, we (human beings) don't play a direct role in setting the weights between units.
(PSSH) Physical Symbol System Hypothesis
a physical symbol system has the necessary and sufficient means for general intelligent action
Intelligence in physical symbol system hypothesis (and two options)
efficiently navigating a search space - option 1: Generate a complete "tree", see which route gets you the solution more quickly -option 2: Do a heuristic search- at each level in the "tree" select the option that gets you closer to the goal
