1. evolutionary view
1.1. why do we have intelligence?
1.2. traditional answer was religious
1.3. modern answer: it is "adaptive", i.e., it helps survival and procreation
1.4. brains are expensive to build and maintain
1.4.1. require rare fats in the diet
1.4.2. require large amounts of energy to maintain
1.5. which capabilities came first?
1.5.1. language?
1.5.2. tool making?
1.5.3. social interaction?
1.5.4. hunting?
1.6. what was the driving factor?
1.6.1. scarcity of food
1.6.2. predators
1.6.3. impressing your mate
1.7. we'll talk about evolution later again
2. animal intelligence
2.1. If we want to understand intelligence, it may be useful to look at how it started in the animal kingdom.
2.2. object recognition, skill learning, etc. are easily seen to be present in many animals
2.3. animal language
2.3.1. attempts at teaching chimps and gorillas human languages
2.3.1.1. sign language (subject to interpretation)
2.3.1.2. plastic tokens and touch screens
2.3.1.3. Kanzi responding to spoken commands
2.3.1.4. but... maybe just "OBJECT LOCATION", no evidence of grammar
2.3.2. attempts at understanding the complexity of chimp, gorilla, dolphine, whale, etc. communications
2.3.2.1. data only now becoming available
2.3.2.2. apply machine learning, statistical modeling for analyzing and decoding it
2.4. tool use
2.4.1. use existing tool
2.4.2. create a tool (e.g., bending a hook)
2.4.3. use one tool to obtain another
2.4.4. crows are capable of doing this
2.4.5. video
2.4.6. bonobos are capable of tool making as well
2.4.7. video
2.5. planning
2.5.1. video game playing
2.5.2. video
2.6. transmission of knowledge
2.6.1. potato washing, separating rice and dirt by floating the rice on water
2.6.2. teaching these skills to the next generation
2.6.3. observed in Japanese snow monkeys
2.6.4. video
2.7. politics
2.7.1. chimps - fighting, alliances, friends, displays, family relations, grooming, etc.
2.7.2. integral part of their society
2.7.3. video
2.8. numeracy
2.8.1. counting objects independent of structure, identity, or arrangement
2.8.2. video
2.9. episodic memory
2.9.1. "Where were you yesterday just before lunch?"
2.9.2. Can we test episodic memory only with language?
2.9.3. Animals remember specific actions and events, e.g., where food items are present / stored.
2.9.4. It is unclear whether this works using the same neural structures as humans or whether it is a different mechanism.
2.10. other examples
2.10.1. learning by observation
2.10.2. cooperation and fairness
2.10.3. problem solving
2.10.4. language
2.10.5. mirror test
2.10.6. non-procreative sex, prostitution
2.10.7. emotion reading
2.10.8. Piaget
2.11. some observations
2.11.1. underlying hypothesis: "these behaviors are so complex that there is no other explanation than complex cognitive processes"
2.11.2. for some of these, we may be mistaken; there may be simple mechanisms that result in seemingly complex behaviors
2.11.2.1. "face on mars"
2.11.2.2. what's actually there
2.11.2.3. we have a tendency to interpret things in human terms
2.11.2.4. "anthropomorphic fallacy"
2.11.3. the "Clever Hans" problem
2.11.3.1. Clever Hans
2.11.3.2. Clever Hans (in German, der Kluge Hans) was an Orlov Trotter horse that was claimed to have been able to perform arithmetic and other intellectual tasks. After a formal investigation in 1907, psychologist Oskar Pfungst demonstrated that the horse was not actually performing these mental tasks, but was watching the reaction of his human observers
2.11.4. on the other hand, a lot of seemingly "intelligent" or "complex" behavior in humans may actually be fairly simple
2.11.5. much of the discussion implicitly assumes that intelligence and language are all/nothing phenomena; but if they evolved gradually over millions of years, we would expect to find simple forms of "intelligence" in many mammals, and it becomes a question of where we draw the line
2.11.6. an inability to learn human language is not the same as ain inability to learn language; animals might be capable of other forms of symbolic communication, or maybe language could be non-symbolic