Differences between revisions 1 and 21 (spanning 20 versions)

Lab Meeting, Autumn 2015, Week 8

This week we are going to talk about distributions. We are going to discuss what it means for motor/speech behavior to change and what that entails for the change in the distribution of that behavior. You can prepare for lab meeting in two ways. Either:

Read this seminal paper about motor movements and accuracy (attached here): Fitts, P. M. (1954). The information capacity of the human motor system in controlling the amplitude of movement. Journal of Experimental Psychology, 47(6), 381–391. doi:10.1037/h0055392

or:
1. Please draw a normal distribution (let's call it A, with a mean µ_A and deviation σ_A).
2. Draw a second distribution, B, with a greater mean value than A (µ_B > µ_A), but equivalent deviation (σ_B = σ_A).
3. Draw a third distribution, C, with the same mean as B (µ_B = µ_C) but a larger deviation (σ_C > σ_A).
4. Notice that our statistical analyses (typically) test whether distributions have equivalent means (e.g. Is µ_A equal to µ_C?). That is, many interpretations of behavior come from interpreting differences in means.
5. Draw a fourth distribution, D, with a mean value equivalent to B and C (µ_D = µ_C = µ_B) but with a positive skew such that its mode is equivalent to the mean of A.
6. Imagine distribution A represent observed motor behavior in an experimental condition.
7. What kinds of conclusions would you draw if an experimental manipulation resulted in a change in behavior such that one observed distribution B? What if instead of B one observed C, would the conclusions be the same? And what if one observed D?
  Here is some R code to generate these distributions if you prefer: https://goo.gl/fr9bxz

Awe- and Aw-inspiring readings (or just stuff you think deserves a mention)

Interesting paper testing whether the timing of childrens' motor movements follow Fitts' law:

Bertucco, M., Bhanpuri, N. H., & Sanger, T. D. (2015). Perceived cost and intrinsic motor variability modulate the speed-accuracy trade-off. PLOS ONE, 10(10), e0139988. http://doi.org/10.1371/journal.pone.0139988 -eb

What we did over last week

Florian

Took a breather from teaching and enjoyed watching Esteban teach. Instead spent a criminal amount of time editing the CrowdExp Tutorial and sending page-long bug reports about CrowdExp to Andrew.
Edited Masha's paper on constituent order and case-marking, which was accepted toCognitive Science. Woohaa.
Proofs for Kuperberg & Jaeger for LCN.
Proofs (final round) for Harris et al special issue on cross-linguistic research in psycholinguistics for LCN.
Met with Jane Grimshaw to revive complementizer project (at the interface of syntax/semantics and online language production)
Met with Carlos Gallo to revive project on inter-clausal production for special issue in Lingua.
Feedback on Wednesday's NSF proposal.
Feedback on Zach's experiment proposal.
Feedback on proposal for special issue in Royal Proceedings - Series B for Damian Blasi and Sean Roberts.
Marvelled at Linda's array of online paradigms for accent perception studies. I think we'll be looking forward to an onslaught of studies on this topic! Hurray.
Declined two requests to review (JML, Frontiers) agreed to one (Cognition).
Reviewed one paper for Cognition.

Kodi

Met with Florian, Chigusa and Maryam about review paper on language processing as inference under uncertainty based on socio-indexical knowledge. We now have a clear shared vision!
Continued writing/revising said review paper
Wrote a bunch of praat scripts to edit stimuli and embed them in noise (for replication/extension of Bradlow and Bent, 2008)
Met with Dave to work on morphing vowel continua with tandemStraight for distributional learning project
Edited stimuli for subtitle project (with Zach and Linda)
Accepted a review invitation from JML

Andrew

BCS 152
BCS 152
BCS 152
BCS 152
BCS 152

Olga

Worked on the tutorial; fixed the lists, redid some css, updated navigation, made screenshots
Trying to figure what happened on the website that it now has a broken link

Esteban

prepped and gave two lectures in BCS 152
conferred with Anne Pier on the design of a eye tracking study
edited materials for the SPR tutorial
re-worked a bunch of distribution analyses for the JML paper
delved into the phonetics literature for (lack of) invariant properties of voicing contrasts

Dave

Wrote (mostly) the introduction to the animals paper
Did final revisions on and submitted a few more job apps.
Ported and improved my analysis code from other fMRI projects for the speech adaptation imaging project, and did some preliminary similarity analyses.
Met with Kodi and walked through morphing vowel continua in STRAIGHT. So far, so good. The repository is under the HLPLab team so check it out if you're curious/want to follow along

Sarah

Continued editing SET manuscript for Lap Phon special issue.
Reviewed stimuli for next Lookslike experiment.
Met with students for BCS 111.
Skyped with Laura Dilley about a potential post-doc.

Dan

Amanda

Learned about using random slope & random intercept mixed effects logistical regressions, and what to do when they don't converge. Spent a lot of time trying to wrap my head around what should be included in the random slope / intercept calculations.
Helped my RA Val figure out the design of her study, which is coupled with:
Tried to get everything set-up on one of the Kinder Lab tablet computers to run eye-tracking tasks with the EyeTribe
Recorded a bunch of people for my new kid task, wound up using my own recordings.
Made some puppet videos for my new study. Contemplating making a few blooper reels in my spare time.
Set-up the new study so that my RAs can run kids this week, and I can tell you about it next week (this will be done by Monday, but is not currently done as of noon on Saturday).
Revised the mammals paper to include the results from the new analyses (see: 1), and to accommodate for reviewer requests. Final draft to be submitted by Tuesday.

Zach

Presented Levy (2008) for BCS 501
Presented for my committee check up meting
Coded Robbie's homework
Met with Wednesday to discuss Robbie's project
Worked on my NSF. I feel like I have great ideas, but not enough time to write it.
Going over the BCS 152 exam with undergrads who took the text
Tested the CrowdExp site for bugs

Linda

Coded up Clarke and Garrett style replication block for the BCS replication class.
Added glossary words that Olga requested to BCS tutorial
Presented in Robbie's class
Met with Carol to discuss Robbie's final project. We're going to be working with Dave's fmri animal dataset.
Currently prepping Bradlow & Bent replication Exp1 for live run! Working on sending out a version for feedback by the end of the weekend, maybe even today.
Continued working on NSF. Goal is to have drafts by the end of the weekend.

Maryam

Wednesday

Spent 5ever on Robbie's homework
Presented Bell et al. (2009) in Chigusa & Mike's class
Talked with Zach about possible final projects for Robbie's class
Brainstormed ideas for other possible projects

-  ⇤ ← Revision 1 as of 2015-08-19 19:06:25 → 
  Size: 503
  Editor: 128
  Comment:
+   ← Revision 21 as of 2015-10-27 14:16:01 → ⇥
  Size: 8296
  Editor: wifi
  Comment:
-Deletions are marked like this.
+Additions are marked like this.
 Line 8:
+This week we are going to talk about distributions. We are going to discuss what it means for motor/speech behavior to change and what that entails for the change in the ''distribution of that behavior''. You can prepare for lab meeting in two ways. Either:

  Read this seminal paper about motor movements and accuracy ([[attachment:Fitts_1954.pdf|attached here]]): Fitts, P. M. (1954). The information capacity of the human motor system in controlling the amplitude of movement. ''Journal of Experimental Psychology'', 47(6), 381–391. [[http://doi.org/10.1037/h0055392|doi:10.1037/h0055392]]

 or:
 
  1. Please draw a normal distribution (let's call it A, with a mean µ,,A,, and deviation σ,,A,,).
  1. Draw a second distribution, B, with a greater mean value than A (µ,,B,, > µ,,A,,), but equivalent deviation (σ,,B,, = σ,,A,,).
  1. Draw a third distribution, C, with the same mean as B (µ,,B,, = µ,,C,,) but a larger deviation (σ,,C,, > σ,,A,,).
  1. Notice that our statistical analyses (typically) test whether distributions have equivalent means (e.g. ''Is µ,,A,, equal to µ,,C,,?''). That is, many interpretations of behavior come from interpreting differences in means.
  1. Draw a fourth distribution, D, with a mean value equivalent to B and C (µ,,D,, = µ,,C,, = µ,,B,,) but with a positive skew such that its mode is equivalent to the mean of A.
  1. Imagine distribution A represent observed motor behavior in an experimental condition.
  1. What kinds of conclusions would you draw if an experimental manipulation resulted in a change in behavior such that one observed distribution B? What if instead of B one observed C, would the conclusions be the same? And what if one observed D?

  Here is some R code to generate these distributions if you prefer: [[https://goo.gl/fr9bxz]]
-Line 9:
+Line 25:
+Interesting paper testing whether the timing of childrens' motor movements follow Fitts' law:
  Bertucco, M., Bhanpuri, N. H., & Sanger, T. D. (2015). Perceived cost and intrinsic motor variability modulate the speed-accuracy trade-off. PLOS ONE, 10(10), e0139988. [[http://doi.org/10.1371/journal.pone.0139988]] -eb
-Line 14:
+Line 33:
-=== Chigusa ===
+. Took a breather from teaching and enjoyed watching Esteban teach. Instead spent a criminal amount of time editing the [[https://www.hlp.rochester.edu/CrowdExp/Tutorial/ | CrowdExp Tutorial]] and sending page-long bug reports about [[https://www.hlp.rochester.edu/CrowdExp/ | CrowdExp]] to Andrew.
 1. '''Edited''' Masha's paper on constituent order and case-marking, which was '''accepted''' to''Cognitive Science''. Woohaa.
 1. '''Proofs''' for Kuperberg & Jaeger for ''LCN''.
 1. '''Proofs''' (final round) for Harris et al special issue on cross-linguistic research in psycholinguistics for ''LCN''.
 1. Met with Jane Grimshaw to revive complementizer project (at the interface of syntax/semantics and online language production)
 1. Met with Carlos Gallo to revive project on inter-clausal production for special issue in ''Lingua''.
 1. '''Feedback''' on Wednesday's NSF proposal.
 1. '''Feedback''' on Zach's experiment proposal.
 1. '''Feedback''' on proposal for special issue in ''Royal Proceedings - Series B'' for Damian Blasi and Sean Roberts.
 1. Marvelled at Linda's array of online paradigms for accent perception studies. I think we'll be looking forward to an onslaught of studies on this topic! Hurray. 
 1. '''Declined''' two requests to review (JML, Frontiers) '''agreed''' to one (Cognition).
 1. '''Reviewed''' one paper for ''Cognition''.


=== Kodi ===
 1. '''Met''' with Florian, Chigusa and Maryam about review paper on language processing as inference under uncertainty based on socio-indexical  knowledge. We now have a clear shared vision!
 1. Continued '''writing/revising''' said review paper
 1. '''Wrote''' a bunch of praat scripts to edit stimuli and embed them in noise (for replication/extension of Bradlow and Bent, 2008)
 1. '''Met''' with Dave to work on morphing vowel continua with tandemStraight for distributional learning project
 1. '''Edited''' stimuli for subtitle project (with Zach and Linda)
 1. Accepted a review invitation from ''JML''
-Line 17:
+Line 56:
+. BCS 152
 2. BCS 152
 3. BCS 152
 4. BCS 152
 5. BCS 152
-Line 19:
+Line 63:
+. Worked on the tutorial; fixed the lists, redid some css, updated navigation, made screenshots
 2. Trying to figure what happened on the website that it now has a broken link
-Line 22:
+Line 68:
+. prepped and gave two lectures in BCS 152
 1. conferred with Anne Pier on the design of a eye tracking study
 1. edited materials for the SPR tutorial
 1. re-worked a bunch of distribution analyses for the JML paper
 1. delved into the phonetics literature for (lack of) invariant properties of voicing contrasts
-Line 24:
+Line 76:
+ * '''Wrote''' (mostly) the introduction to the animals paper
 * Did final revisions on and submitted a few more '''job''' apps.
 * Ported and improved my '''analysis''' code from other fMRI projects for the [[https://bitbucket.org/dkleinschmidt/speech-adapt-fmri|speech adaptation imaging project]], and did some preliminary similarity analyses.
 * Met with Kodi and walked through '''morphing''' vowel continua in STRAIGHT.  So far, so good.  The [[https://bitbucket.org/hlplab/supunsup-vowels|repository]] is under the HLPLab team so check it out if you're curious/want to follow along :)
-Line 25:
+Line 82:
+. Continued '''editing''' SET manuscript for Lap Phon special issue.
 1. Reviewed stimuli for next Lookslike experiment.
 1. Met with students for BCS 111.
 1. Skyped with Laura Dilley about a potential post-doc.
-Line 29:
+Line 90:
+. '''Learned''' about using random slope & random intercept mixed effects logistical regressions, and what to do when they don't converge. Spent a lot of time trying to wrap my head around what should be included in the random slope / intercept calculations.
 1. '''Helped''' my RA Val figure out the design of her study, which is coupled with:
 1. Tried to get everything '''set-up''' on one of the Kinder Lab tablet computers to run eye-tracking tasks with the [[http://www.pygaze.org/2015/06/eyetribe/|EyeTribe]]
 1. '''Recorded''' a bunch of people for my new kid task, wound up using my own recordings. 
 1. '''Made''' some puppet videos for my new study. Contemplating making a few blooper reels in my spare time.
 1. '''Set-up''' the new study so that my RAs can run kids this week, and I can tell you about it next week (this will be done by Monday, but is not currently done as of noon on Saturday).
 1. '''Revised''' the mammals paper to include the results from the new analyses  (see: 1), and to accommodate for reviewer requests. Final draft to be submitted by Tuesday.
-Line 31:
+Line 99:
+. '''Presented''' Levy (2008) for BCS 501
 1. '''Presented''' for my committee check up meting
 1. '''Coded''' Robbie's homework
 1. '''Met''' with Wednesday to discuss Robbie's project
 1. Worked on my '''NSF'''. I feel like I have great ideas, but not enough time to write it.
 1. Going over the '''BCS 152 exam''' with undergrads who took the text
 1. '''Tested''' the CrowdExp site for bugs
-Line 33:
+Line 109:
+. Coded up Clarke and Garrett style replication block for the BCS replication class.
 1. Added glossary words that Olga requested to BCS tutorial
 1. Presented in Robbie's class
 1. Met with Carol to discuss Robbie's final project. We're going to be working with Dave's fmri animal dataset.
 1. Currently prepping Bradlow & Bent replication Exp1 for live run! Working on sending out a version for feedback by the end of the weekend, maybe even today.
 1. Continued working on NSF. Goal is to have drafts by the end of the weekend.
-Line 37:
+Line 119:
+. Spent 5ever on Robbie's homework
 1. Presented Bell et al. (2009) in Chigusa & Mike's class
 1. Talked with Zach about possible final projects for Robbie's class
 1. Brainstormed ideas for other possible projects