CS295J/Class Members' Pages/Gideon/Week 1
Literature Keys
- Clark-1998-TEM
- Poster, Owner, Evaluator
- Summary
- Scholl-2000-PCA
- Discussant
- Adamczyk-2004-NNW
- Discussant
- HermerVazquez-1999-SFC
- Poster, Evaluator
- Evans-2003-ITM
- Poster, Evaluator
Summaries
Clark-1998-TEM
The Extended Mind
We should not limit our notion of cognition to something taking place only within the skull. Instead, we can think about actively incorporating the external environment (or being incorporated into it), forming a single-system loop of cognition. To help understand this in terms of how we might utilize a computer's resources, consider the following:
- A beginner tetris player will often "unnecessarily" rotate pieces and move them about before deciding (or being forced to decide) on a final orientation.
- An intermediate tetris player will make fewer "wasteful" moves because he/she will have had the training and/or metal power to perform these simple calculations in his/her head.
- It's not unreasonable to assume an advanced player will take this one step further, achieving a perfect state of efficiency in play, only moving when absolutely necessary.
However, this is not the case! In fact, an advanced player's actions, in many ways, resemble those of a complete beginner. Namely, the expert WILL make moves which, though seemingly unnecessary, are taking advantage of the computers ability to do the same thing he/she would be doing, but better.
Concretely...
| Action | time [ms] |
| Plan a button press | 200 |
| Computerized piece rotation | 100 |
| Mental rotation | 1000 |
Using our environment for cognitive tasks is something we do all the time. Consider the problem of following directions to get to a new restaurant.
- 1) We can write directions down on paper.
- 2) We can map them on our iPhones.
- 3) We can insert them into our neural implants which integrate with memory and spatial navigation systems.
This paper argues that these cases are fundamentally equivalent in terms of distribution of cognition.
People often subconsciously make use of computer resources, and knowing which ones and to what degree will help us predict how users will interact with a system.