= 27d567bc2d48d9f40e9ccc20ea370b15)
beoShaffer comments on Attention control is critical for changing/increasing/altering motivation - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (98)
How does the research determine if the subjects, particularly the animals, are focusing on a task rather than doing it autonomously? In the monkey example it seems like it is based off of the complexity of the task, but if that is the case it seems like the effects could be from complex tasks(which may be done autonomously) rather than focus.
By altering the reward condition.
In the experimental condition, the animal gets the reward only if it lets go when it feels the speed and vibration of the wheel change in a certain way. So, if it's not paying attention, it doesn't need when to let go.
In the control condition, reward comes at random intervals, regardless of the speed and vibration of the wheel. Since there is no incentive to pay attention to it, the animal habituates quickly and ignores the wheel.
I think you're probably right, but I just wanted to check on the interpretation of this piece of evidence...
In the monkey experiment, how do we know that the differences are due to attention as opposed to just being a direct result of the altered reward condition?
If I had a pattern matching circuit that was driven by reinforcement learning, I would expect it to do nothing meaningful if it was rewarded randomly, and I'd expect it to get better at matching patterns if it gets rewarded when it matches them correctly. Do we even need a notion of "attention" here?
I think the CBT and meditation examples are more convincing to me as with those it's the attention itself that's the variable under control.