Timeline for What experiment in a simple decision task should we run to obtain one million trials?
Current License: CC BY-SA 3.0
34 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 7, 2017 at 0:20 | vote | accept | Denis Cousineau | ||
| Feb 14, 2017 at 1:22 | answer | added | StrongBad | timeline score: -1 | |
| Feb 13, 2017 at 10:02 | answer | added | andyw | timeline score: 1 | |
| Feb 13, 2017 at 9:31 | answer | added | andyw | timeline score: 1 | |
| Feb 12, 2017 at 5:38 | history | tweeted | twitter.com/StackCogSci/status/830652177874092032 | ||
| Feb 12, 2017 at 0:22 | answer | added | Jeff Miller | timeline score: 4 | |
| Feb 8, 2017 at 4:33 | answer | added | Andrew Heathcote | timeline score: 4 | |
| Feb 7, 2017 at 21:04 | answer | added | Scott Brown | timeline score: 2 | |
| Feb 7, 2017 at 19:49 | answer | added | Pierre Jolicoeur | timeline score: 2 | |
| Feb 4, 2017 at 2:59 | comment | added | Jeromy Anglim | @AliceD Thanks for the openness. | |
| Feb 3, 2017 at 10:08 | comment | added | AliceD♦ | @JeromyAnglim - Thanks for looking into my flags. I'll leave my comments here, but if you support this question, feel free to remove my (kind of negative) stuff here if you consider it nothing but clutter. Cheers. | |
| Feb 3, 2017 at 1:26 | comment | added | Jeromy Anglim | I've given it a tiny edit for structure. I think the final bit "Please let's see what are..." would normally not be required on this site. But it appears Denis is inviting new researchers to the site to contribute (which is fantastic), so I think these notes are still useful. | |
| Feb 3, 2017 at 1:17 | history | edited | Jeromy Anglim | CC BY-SA 3.0 |
added 149 characters in body
|
| Feb 3, 2017 at 1:12 | comment | added | Jeromy Anglim | I think this is a very different kind of question. It does go against some of the rules of the site. It could be adapted further to the norms of the site. That said, Denis (as a senior academic), I would like to welcome you to the site. My preference would be to give this question some latitude, given that we are keen to engage more academics and active researchers on the site. Perhaps we could propose and make edits to the question to bring the question more in line with expectations. | |
| Feb 2, 2017 at 21:02 | answer | added | user14603 | timeline score: 4 | |
| Feb 1, 2017 at 19:43 | answer | added | Marc-André Goulet | timeline score: 4 | |
| Feb 1, 2017 at 14:23 | history | edited | Denis Cousineau | CC BY-SA 3.0 |
Added other benefits of such data set.
|
| Feb 1, 2017 at 9:50 | comment | added | AliceD♦ | See the help-center: [... open-ended questions diminish the usefulness of our site and push other questions off the front page ].Go to the chat room for chatty conversations. Go to the meta site to find out whether a type of question is appropriate. | |
| Feb 1, 2017 at 0:24 | history | edited | Denis Cousineau | CC BY-SA 3.0 |
Added the fact that the choice of precise task is independent of the aim of the project.
|
| Jan 31, 2017 at 16:09 | answer | added | user14888 | timeline score: 3 | |
| Jan 31, 2017 at 16:00 | comment | added | Denis Cousineau | See meta What is cognitive sciences about for my view on this. | |
| Jan 31, 2017 at 15:54 | history | edited | Denis Cousineau | CC BY-SA 3.0 |
remove reference to the fact that the winning task will actually be run. Was perceived as a contest by some moderators...
|
| Jan 31, 2017 at 15:42 | history | edited | Denis Cousineau | CC BY-SA 3.0 |
Better define what the "best task" would consist of
|
| Jan 31, 2017 at 8:32 | comment | added | AliceD♦ | might be but you may benefit from reading the help center on asking proper questions. The idea can be OK, the platform you've chosen to execute it is not. | |
| Jan 31, 2017 at 2:01 | answer | added | ElenorDavey | timeline score: -1 | |
| Jan 30, 2017 at 23:28 | comment | added | Denis Cousineau | It is not a competition, it is a collaborative project. Getting a lot of researchers in experimental psychology to brainstorm alternatives is definitely the best way to find the optimal task. | |
| Jan 30, 2017 at 22:28 | comment | added | AliceD♦ | This is a post begging for some kind of pob competition. This is definitely not the site to start out some kind of in silico award for the best psychophysical task. -1 & vote-to-close & flagging too on top of it. | |
| Jan 30, 2017 at 21:25 | review | Close votes | |||
| Feb 12, 2017 at 3:02 | |||||
| Jan 30, 2017 at 19:19 | history | edited | Denis Cousineau | CC BY-SA 3.0 |
added 21 characters in body
|
| Jan 30, 2017 at 19:15 | comment | added | Robin Kramer-ten Have | Good point. I think it would be interesting to have a lot of different paradigms gathered here. As for the discussion part, that should be started in a chat, as the comments/answers are no place for discussions. | |
| Jan 30, 2017 at 19:13 | comment | added | Denis Cousineau | As of the label "best", that would be with regard to those criteria: small number of response alternatives, few conditions, low error rates, stimuli that can be characterized for modeling purposes, possibility of extended training on that task. Hence,with these objective criteria in mind, the question is no longer opinion-based. | |
| Jan 30, 2017 at 19:11 | comment | added | Denis Cousineau | @RobinKramer: The current models of cognitions (sampling models such as LBA or Diffusion) are very flexible; they aim at predicting performance in a wide variety of situations. Hence, the exact task used is not a critical constraint when testing these models. For example, they can be tested on lexical decision tasks, dot motion tasks, numerosity tasks, matching tasks, etc., all tasks that are very different but make similar assumptions about the underlying processes (accumulation of evidence+thresholds). However, what is critical is the sample size; with small $n$, all models offer good fits. | |
| Jan 30, 2017 at 19:02 | comment | added | Robin Kramer-ten Have | I like the idea, but asking for "the best" is rather opinion based, which is a reason for closure of the question. Could you explain a little bit more about what you are researching and why you (and others) want such a large amount of data? Don't cognitive models, and thus the data used, depend on the specific research questions? | |
| Jan 30, 2017 at 17:31 | history | asked | Denis Cousineau | CC BY-SA 3.0 |