Week 4 Flashcards
(21 cards)
Cognition (Dr. Johnson)
adaptive engagement with the world
in the ways animals interact, change and engage with their environment
sensory motor constraints
kind of access to o the world an animal has
Bottlenose dolphins
sees world in shades of gray :(
They do not do well with sample matches involving visual stimuli, but excel at auditory stimuli.
Dolphin vision
sensitive to motion and high contrast
no fovea in the eye so can’t distinguish color
Primates (inc human) have…
- forward facing eyes. (good depth perception)
- Hunting/grabbing and arboreal locomotion/climbing - Opposable thumbs, grasping hands
Hand-eye coordination
primates can see their own hand
-human (& chimps) engage in tool
-tend to have more elaborate mind
Other “hand-eye” coordination
elephant: trunk-eye coordination
raven: eye-beak-foot coordination
General Principle of ALL animals
- detect and use event correlations. use event 1 to predict event 2 (ex. pavlov’s dog)
- Win stay/Lose shift
general principle can be overridden by ecological. demands (ex: humming bird win/shift is better than win/stay)
Temporal Continuity
Ex: Pavlov’s dog: Bell & Food must occur close in time for animal to learn correlation
What do we need to keep in mind when designing cognitive experiments?
Ecological validity
Humans have hand-eye coordination. Elephants have trunk-eye coordination. Ravens have foot-eye-mouth coordination. These are species-specific sensory-motor abilities which best illustrate which of the following?
Emergence
Embodiment
Cognitive ecology
Distributed cognition
embodiment
Taste Aversion Learning (rodents only?)
animal realizes that it will take them some time to feel negative effects from the food they are eating
- their brain recognizes that it would take an hour
after eating to feel physical effects
Dr. Johnson pointed out that rats can find a reward in a maze, regardless of their starting position. This suggests that they have formed a ‘mental map’ of the maze. At what level of analysis does this reasoning operate at?
Algorithmic
Perspective Encoding
Lines interfere late, colors interfere early
Symbol use
once number trained, chimps are better at number recognition task than humans.
Why data science?
Collect data to make predictions.
Jobs have high satisfaction rate
Sexiest job or smth like that
Data engineer vs scientist
E: builds system that can ingest data from multiple sources which one can do analysis on.
S: write and run code that can do statistical analysis and draws predictions from data.
Why the trouble defining field?
People confuse the essence of what you’re doing with the tools that you use.
BrainSCAN
- used to find patterns in research papers
Based on statistical patterns
Hypothesis generation algorithm
UBER
- extracted patterns from historical data
- compared temporal and spatial profiles of different neighborhoods
- analyze seemingly unrelated data to improve your system and make service better
Neural Data Science (future?)
We can look for connects between cell types, gene expression, function, oscillations, connectivity to see how they all interrelate.
Find gaps and connect to find missing links.
Use data to generate hypotheses.
