I think esp. with regard to data collection awareness is needed (bit more optimistic about processing). Also of the implications: hypothesis tests in complete isolation are impossible since we're also testing (at the very least) theoretical considerations behind the measurements.
-
-
Agreed on the 2nd point! Re: 1st point: there my be differences between our respective disciplines, and its main purposes, eg treatment vs understanding.
2 replies 0 retweets 1 like -
I agree, although in this case I meant to refer to understanding (and a bit to your field in which I'm greatly interested). My impression was that there's a growing awareness of the role of theory in data processing (in comparison to data coll.), but you seem more sceptical?
1 reply 0 retweets 2 likes -
Depends on what I consider my field. If I think of it as ‘(computational) cognitive science’, then yes. However, if I think of it as ‘psychology’, then I think we still have very far to go: too much focus on data & too little theory, nor much reflection on theory-data interface.
2 replies 1 retweet 7 likes -
Yes!
1 reply 0 retweets 2 likes -
Iris van Rooij Retweeted 🐙 🛐Will Gervais 🛐 🐙
Here is a relevant poll, plus discussion: https://twitter.com/wgervais/status/957319964347764736 … (specific for social psychology, but I’m not sure eg developmental & cogn psychology would score much differently, unless the latter are part of cogsci).
Iris van Rooij added,
1 reply 2 retweets 3 likes -
Replying to @IrisVanRooij @o_guest and
does the use of modeling factor in here? do you think areas with more of a computational tradition tend to have more specific theoretical predictions?
1 reply 1 retweet 2 likes -
Replying to @PratherLab @o_guest and
Indeed. I see computational modelling as theoretical tool, so yes that factors in. For me personally, primary role of theory/modeling is understanding and explanation, and prediction is secondary (means to an end, not the end itself as some psychologists seem to phrase/view it).
1 reply 2 retweets 5 likes -
Replying to @IrisVanRooij @PratherLab and
Agreed. Modelling in comp cog/neuro is primarily explanatory although often predictions fall out of the explanatons. This is then fed back into the overarching theoretical framework which might then go on to make predictions.
2 replies 3 retweets 4 likes -
Replying to @o_guest @IrisVanRooij and
Depends who's creating the papers. My lab puts untested predictions in the model papers. It's a great way to write and ensures ( baring outright lying) that predictions are not post hoc.
1 reply 0 retweets 1 like
I don't think what we said precludes that at all.
-
-
Replying to @o_guest @IrisVanRooij and
I interpreted "fall out" as meaning accidental rather than deliberate but I can see you might have meant it differently.
0 replies 0 retweets 1 likeThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.