It's Good Friday.
As a sacrifice to Science, I'm going to criticise every single paper I've ever written, one tweet at a time.
Stay tuned..!
#ImAVeryNaughtyBoy
As a sacrifice to Science, I'm going to criticise every single paper I've ever written, one tweet at a time.
Stay tuned..!
#ImAVeryNaughtyBoy
Paper #1: Holmes & Spence 2004 @xmodal
Criticism: A non-systematic review of selected papers, no assessment of study quality, bias, or effect sizes. The only novel thing in this paper was a long, harsh critique of Iriki et al. (1996)
#SelfCriticism
doi.org/10.1007/s10339…
Criticism: A non-systematic review of selected papers, no assessment of study quality, bias, or effect sizes. The only novel thing in this paper was a long, harsh critique of Iriki et al. (1996)
#SelfCriticism
doi.org/10.1007/s10339…
Paper #2: Skaliora et al. 2004
Criticism: I contributed only one figure (#8) to this paper, but >30 neonatal rats paid the ultimate price for my crappy patch-clamping skills (only 17 cells) as an MSc student. I still feel the guilt.
#SelfCriticism
doi.org/10.1152/jn.004…
Criticism: I contributed only one figure (#8) to this paper, but >30 neonatal rats paid the ultimate price for my crappy patch-clamping skills (only 17 cells) as an MSc student. I still feel the guilt.
#SelfCriticism
doi.org/10.1152/jn.004…
Paper #3: Holmes et al. 2004
Criticism: E1 (N=10) was a BSc student's project which I doubted when Charles asked me to write it up. After peer review I begrudgingly replicated it (but only N=6). I was surprised it worked. the p-values are poorly reported
doi.org/10.3758/CABN.4…
Criticism: E1 (N=10) was a BSc student's project which I doubted when Charles asked me to write it up. After peer review I begrudgingly replicated it (but only N=6). I was surprised it worked. the p-values are poorly reported
doi.org/10.3758/CABN.4…
Paper #4: Spence et al. 2004
Criticism: Another non-systematic review. I learnt a lot more from reading this than I contributed. I feel very lucky to have picked a superstar for my PhD supervisor - he co-opted me in to half of my papers in that time.
doi.org/10.1016/j.jphy…
Criticism: Another non-systematic review. I learnt a lot more from reading this than I contributed. I feel very lucky to have picked a superstar for my PhD supervisor - he co-opted me in to half of my papers in that time.
doi.org/10.1016/j.jphy…
Paper #5: Holmes et al. 2004
Criticism: After rejection from multiple 'high-tier' journals (Nat Neurosci, Curr Biol 🤣), this 1-expt, N=16 paper was enlarged to 3 expts, N=24 each. The paper (& my career...) relies on a 3-way interaction, p=.045. Weak.
doi.org/10.1016/j.neul…
Criticism: After rejection from multiple 'high-tier' journals (Nat Neurosci, Curr Biol 🤣), this 1-expt, N=16 paper was enlarged to 3 expts, N=24 each. The paper (& my career...) relies on a 3-way interaction, p=.045. Weak.
doi.org/10.1016/j.neul…
Paper #6: Ehrsson et al. 2005
Criticism: I'm an author because my stand-in supervisor for a few months (thanks @Dick_retired!), suggested I help Henrik @BrainSelf with his experiment. I pushed 'start' in the control room, while Henrik worked. Lucky move!
doi.org/10.1523/JNEURO…
Criticism: I'm an author because my stand-in supervisor for a few months (thanks @Dick_retired!), suggested I help Henrik @BrainSelf with his experiment. I pushed 'start' in the control room, while Henrik worked. Lucky move!
doi.org/10.1523/JNEURO…
Paper #7: Holmes et al. 2005
Criticism: The data & analysis are fine, but it's much too reliant on ANOVA. It would be much better to fit linear models to each participant's data & do all inferential stats on the model fits. Need to show individual data!
doi.org/10.1007/s00221…
Criticism: The data & analysis are fine, but it's much too reliant on ANOVA. It would be much better to fit linear models to each participant's data & do all inferential stats on the model fits. Need to show individual data!
doi.org/10.1007/s00221…
Paper #8: Holmes & Spence 2005
Criticism: I actually love this one, but there's a typo on p763, 1st col: "was" should be "wasn't". It has bugged me forever. In writing this paper I started to realise that the law of #InverseEffectiveness was a #DoubleDip
doi.org/10.1016/j.cub.…
Criticism: I actually love this one, but there's a typo on p763, 1st col: "was" should be "wasn't". It has bugged me forever. In writing this paper I started to realise that the law of #InverseEffectiveness was a #DoubleDip
doi.org/10.1016/j.cub.…
Paper #9: Holmes et al. 2006
Criticism: I was going to do the right thing & report individual linear data & fits (+ ANOVA & t-tests), but a single 'outlier' participant in Expt 2 made the stats 'non-significant'. I reverted to ANOVA: tail wagged the dog.
doi.org/10.3758/BF0320…
Criticism: I was going to do the right thing & report individual linear data & fits (+ ANOVA & t-tests), but a single 'outlier' participant in Expt 2 made the stats 'non-significant'. I reverted to ANOVA: tail wagged the dog.
doi.org/10.3758/BF0320…
Paper #10: Holmes et al. 2006
Criticism: In the lab we always looked at reaction times (RT) errors (%) & their combination 'inverse efficiency', IE=RT/p(correct). Papers relied either on RT, % or IE. This paper used % because the p-value for IE was 0.06
doi.org/10.1016/j.brai…
Criticism: In the lab we always looked at reaction times (RT) errors (%) & their combination 'inverse efficiency', IE=RT/p(correct). Papers relied either on RT, % or IE. This paper used % because the p-value for IE was 0.06
doi.org/10.1016/j.brai…
Paper #11: Snijders et al. 2007
Criticism: In the series of 'mirror illusion' papers, this is the best one: clear hypotheses, individual data presented, all stats shown, & no fudging (AFAI recall...). Credit must go to Dirk's MSc project rather than me.
doi.org/10.1016/j.neur…
Criticism: In the series of 'mirror illusion' papers, this is the best one: clear hypotheses, individual data presented, all stats shown, & no fudging (AFAI recall...). Credit must go to Dirk's MSc project rather than me.
doi.org/10.1016/j.neur…
Papers #12 & #13: Holmes (2007, 2009)
Criticism: Each of these made the same, relatively small, point about a particular paper or finding that I wanted to criticise. In retrospect, these were just little stepping-stones #DoubleDip
doi.org/10.1016/j.conc…
doi.org/10.1016/j.neur…
Criticism: Each of these made the same, relatively small, point about a particular paper or finding that I wanted to criticise. In retrospect, these were just little stepping-stones #DoubleDip
doi.org/10.1016/j.conc…
doi.org/10.1016/j.neur…
Paper #14: Makin et al. 2007
Criticism: I was not involved in data collection/analysis, instead just discussion & writing. The paper relies on a 2x4 statistical interaction, but doesn't explicitly test it. I wanted to check this using FSL, but didn't...
doi.org/10.1523/JNEURO…
Criticism: I was not involved in data collection/analysis, instead just discussion & writing. The paper relies on a 2x4 statistical interaction, but doesn't explicitly test it. I wanted to check this using FSL, but didn't...
doi.org/10.1523/JNEURO…
Paper #15: Holmes & Spence 2007
Criticism: This should have been a thoughtful commentary on @Dijkermanlabs massive Behav Brain Sci paper on somatosensory processing streams, but I used it to make my own points about the rubber hand illusion. Gratuitous.
doi.org/10.1017/S01405…
Criticism: This should have been a thoughtful commentary on @Dijkermanlabs massive Behav Brain Sci paper on somatosensory processing streams, but I used it to make my own points about the rubber hand illusion. Gratuitous.
doi.org/10.1017/S01405…
Paper #16: Holmes et al. 2007
Criticism: At 15000 words, this egregiously-long paper was my attempt to show that #ToolUse affects multisensory spatial attention. Part data-dump, part abyss-shout, I should have respected the reader's time a LOT more.
doi.org/10.1016/S0010-…
Criticism: At 15000 words, this egregiously-long paper was my attempt to show that #ToolUse affects multisensory spatial attention. Part data-dump, part abyss-shout, I should have respected the reader's time a LOT more.
doi.org/10.1016/S0010-…
Paper #17: Makin et al. 2008
Criticism: A selective review & model of how the #RubberHandIllusion relates to #PeripersonalSpace. I still remember the moment I suggested, quite late in the writing process, "Should we add a model?" #PostHoc #ModelHacking
doi.org/10.1016/j.bbr.…
Criticism: A selective review & model of how the #RubberHandIllusion relates to #PeripersonalSpace. I still remember the moment I suggested, quite late in the writing process, "Should we add a model?" #PostHoc #ModelHacking
doi.org/10.1016/j.bbr.…
Paper 18: Holmes et al. 2008
Criticism: My first #fMRI paper. Design was not optimal; was based too-much on my behavioural expts. The effects are weak, the whole-brain analysis failed & it relied on region-of-interest analysis. I doubt it would replicate
doi.org/10.1371/journa…
Criticism: My first #fMRI paper. Design was not optimal; was based too-much on my behavioural expts. The effects are weak, the whole-brain analysis failed & it relied on region-of-interest analysis. I doubt it would replicate
doi.org/10.1371/journa…
Paper 19: Bremner et al. 2008 @Andy_Bremner
Criticism: I honestly can not remember either what is in this paper or what my contribution was. I doubt I could pick-out my contribution(s?) in a two-alternative forced-choice design. Sorry. This is a cop-out!
doi.org/10.1016/j.tics…
Criticism: I honestly can not remember either what is in this paper or what my contribution was. I doubt I could pick-out my contribution(s?) in a two-alternative forced-choice design. Sorry. This is a cop-out!
doi.org/10.1016/j.tics…
Paper 20: Makin et al. 2009
Criticism: Still a favourite, this paper comes with 2 warnings from me: The selection of trials for inclusion (EMG artefacts) should be automated; We 'tuned' the analysis for E1 data, so the effect-size is over-estimated there
doi.org/10.1523/jneuro…
Criticism: Still a favourite, this paper comes with 2 warnings from me: The selection of trials for inclusion (EMG artefacts) should be automated; We 'tuned' the analysis for E1 data, so the effect-size is over-estimated there
doi.org/10.1523/jneuro…
Paper 21: Holmes 2009
Criticism: In case you didn't believe my first three papers on the topic, here are too many supplementary graphs & simulations making the same, relatively small, statistical point. The point is correct, but it's a paragraph, tops.
doi.org/10.1007/s10548…
Criticism: In case you didn't believe my first three papers on the topic, here are too many supplementary graphs & simulations making the same, relatively small, statistical point. The point is correct, but it's a paragraph, tops.
doi.org/10.1007/s10548…
Paper #22: Revol et al. 2009
Criticism: I am undeservedly a 'middle' author on this paper. These days, I would not accept authorship based on such a small contribution. I would happily defend the paper though!
doi.org/10.1007/s00221…
Criticism: I am undeservedly a 'middle' author on this paper. These days, I would not accept authorship based on such a small contribution. I would happily defend the paper though!
doi.org/10.1007/s00221…
Paper #23: Makin et al. 2012
Criticism: Another non-systematic, selective review. This one puts our TMS work (Paper #20) in a wider context. (Systematic) reviews should be done to answer questions, rather than to interpret and/or promote our own work.
doi.org/10.1007/s00221…
Criticism: Another non-systematic, selective review. This one puts our TMS work (Paper #20) in a wider context. (Systematic) reviews should be done to answer questions, rather than to interpret and/or promote our own work.
doi.org/10.1007/s00221…
Paper #24: Holmes et al. (2007) (sorry, missed one!)
Criticism: The two experiments in this paper had N=30 each, yet all our other similar papers had N=24. Why? We added 2x6 participants after checking the data because the interaction was not significant
doi.org/10.1007/s00221…
Criticism: The two experiments in this paper had N=30 each, yet all our other similar papers had N=24. Why? We added 2x6 participants after checking the data because the interaction was not significant
doi.org/10.1007/s00221…
Paper #25: Holmes 2012
Criticism: This is where my self-criticism really begins. As I wrote in this paper, re-analysis of my own work (Papers #5, #16, & #24), showed that, overall, the effect size we studied was too small to be worth any further study.
doi.org/10.1007/s00221…
Criticism: This is where my self-criticism really begins. As I wrote in this paper, re-analysis of my own work (Papers #5, #16, & #24), showed that, overall, the effect size we studied was too small to be worth any further study.
doi.org/10.1007/s00221…
Paper #26: Naish et al. 2013
Criticism: Great to work with @KatherineNaish & @ArranReader. The problem in E2 is: We can't be sure there was any *signal* for observers to detect - people couldn't detect it, but maybe it wasn't there? Thanks to: @ag3dvr
doi.org/10.1007/s00221…
Criticism: Great to work with @KatherineNaish & @ArranReader. The problem in E2 is: We can't be sure there was any *signal* for observers to detect - people couldn't detect it, but maybe it wasn't there? Thanks to: @ag3dvr
doi.org/10.1007/s00221…
Paper #27: Wilf et al. 2013
Criticism: This was beautifully designed & executed by @MeytalWilf, with me only advising & coding. Looking back, I think the paper would be stronger with a larger sample and/or replication in the healthy participant datasets
doi.org/10.3389/fpsyg.…
Criticism: This was beautifully designed & executed by @MeytalWilf, with me only advising & coding. Looking back, I think the paper would be stronger with a larger sample and/or replication in the healthy participant datasets
doi.org/10.3389/fpsyg.…
Paper #28: Tamè et al. 2014
Criticism: This was my first study on #touch (but not @TameLuigi's!) & it was my first attempt measuring thresholds. We discarded 10/28 participants in E1 - couldn't feel the stimuli. I'm still plagued by this. Touch is hard!
doi.org/10.3389/fpsyg.…
Criticism: This was my first study on #touch (but not @TameLuigi's!) & it was my first attempt measuring thresholds. We discarded 10/28 participants in E1 - couldn't feel the stimuli. I'm still plagued by this. Touch is hard!
doi.org/10.3389/fpsyg.…
Paper #29: Graham et al. 2014
Criticism: I was a long-distance advisor on this project (I've never met my co-authors!). I can't really critique the clinical parts, but there are too many behavioural tasks being used as a 'battery' here. Needs more focus!
doi.org/10.3389/fpsyt.…
Criticism: I was a long-distance advisor on this project (I've never met my co-authors!). I can't really critique the clinical parts, but there are too many behavioural tasks being used as a 'battery' here. Needs more focus!
doi.org/10.3389/fpsyt.…
Paper #30: Naish et al. 2014
Criticism: Planned & submitted as a selective review, our reviewer told us to make it #systematic. My heart sank, but @KatherineNaish rose to the challenge. So much work! Since this paper, it's #SysRev or no review at all.
doi.org/10.1016/j.neur…
Criticism: Planned & submitted as a selective review, our reviewer told us to make it #systematic. My heart sank, but @KatherineNaish rose to the challenge. So much work! Since this paper, it's #SysRev or no review at all.
doi.org/10.1016/j.neur…
Paper #31: Makin et al. 2015
Criticism: This should have worked, but as the most complicated experiment I've been part of (3 experimenters, 2 rubber hands, 1 pug, TMS, LCD goggles, 2 lasers & a small hot room in France), I'm still not sure what we found
doi.org/10.1016/j.cort…
Criticism: This should have worked, but as the most complicated experiment I've been part of (3 experimenters, 2 rubber hands, 1 pug, TMS, LCD goggles, 2 lasers & a small hot room in France), I'm still not sure what we found
doi.org/10.1016/j.cort…
Paper #32: Graham et al. 2015
Criticism: As 'just' an advisor on this project, in retrospect I might question whether N=48 is sufficient for a confirmatory PCA expecting 4 components. That said, my co-authors use R, so they probably know better :-)
doi.org/10.3758/s13414…
Criticism: As 'just' an advisor on this project, in retrospect I might question whether N=48 is sufficient for a confirmatory PCA expecting 4 components. That said, my co-authors use R, so they probably know better :-)
doi.org/10.3758/s13414…
Paper #33: Naish & Holmes 2015
Criticism: My colleague said he'd 'never heard of' the journal this commentary is in. Neither had I. It was a worthwhile exercise with @KatherineNaish, but has anyone actually read it? Commentaries should be easier & opener
doi.org/10.1016/j.plre…
Criticism: My colleague said he'd 'never heard of' the journal this commentary is in. Neither had I. It was a worthwhile exercise with @KatherineNaish, but has anyone actually read it? Commentaries should be easier & opener
doi.org/10.1016/j.plre…
Paper #34: Holmes & Dakwar 2015
Criticism: I literally cannot tell you how many times & in how many ways these data were analysed. Analysis of kinematic data feels like a 'Wild West' - no standard methods. I learnt a LOT, but this is an exploratory paper
doi.org/10.1016/j.visr…
Criticism: I literally cannot tell you how many times & in how many ways these data were analysed. Analysis of kinematic data feels like a 'Wild West' - no standard methods. I learnt a LOT, but this is an exploratory paper
doi.org/10.1016/j.visr…
Paper #35: Reader & Holmes 2015
Criticism: @ArranReader has been a fantastic scientist & colleague over years. In re-visiting this work, I feel I flapped around thinking of ways to analyse these data. I don't doubt the conclusion, but how we got there!
doi.org/10.3389/fpsyg.…
Criticism: @ArranReader has been a fantastic scientist & colleague over years. In re-visiting this work, I feel I flapped around thinking of ways to analyse these data. I don't doubt the conclusion, but how we got there!
doi.org/10.3389/fpsyg.…
Lolz! This paper is in my all-time top-3
BUT: The time-window analysis shown in Figure 4 is a fudge. The stats are OK, but I picked the filter settings that *looked* best
@TameLuigi & I spent much time & many birra thinking about this paper & its 'story'. It's not finished...
BUT: The time-window analysis shown in Figure 4 is a fudge. The stats are OK, but I picked the filter settings that *looked* best
@TameLuigi & I spent much time & many birra thinking about this paper & its 'story'. It's not finished...
Paper #37: Reader & Holmes 2016
Criticism: It's not a #SystematicReview
I really valued narrative or selective reviews like this one when I was a PhD & post-doc, so they must have some use! This was a great paper for @ArranReader to influence the field.
doi.org/10.1007/s40167…
Criticism: It's not a #SystematicReview
I really valued narrative or selective reviews like this one when I was a PhD & post-doc, so they must have some use! This was a great paper for @ArranReader to influence the field.
doi.org/10.1007/s40167…
Paper #38: Tamè et al. 2016
Criticism: Ditto - not #Systematic! I think there should be a clear distinction or grading system for reviews, pointing out which are more 'opinion' & which are more like novel empirical datasets.
doi.org/10.1080/026432…
Criticism: Ditto - not #Systematic! I think there should be a clear distinction or grading system for reviews, pointing out which are more 'opinion' & which are more like novel empirical datasets.
doi.org/10.1080/026432…
Papers #39 & #40: Graham et al. 2016a, b
Criticism: In general with Kyran's work (see Papers #29 & #32) I think my role here was an 'advisor' rather than 'author'. On reflection I am not familiar enough with his work. That's on me
doi.org/10.1016/j.conc…
doi.org/10.1080/135468…
Criticism: In general with Kyran's work (see Papers #29 & #32) I think my role here was an 'advisor' rather than 'author'. On reflection I am not familiar enough with his work. That's on me
doi.org/10.1016/j.conc…
doi.org/10.1080/135468…
Paper #41: McGlashan et al. 2017
Criticism: The experimental group did online movement activities for several weeks, but the controls did nothing. The benefits of intervention need narrowing-down in further studies. (This was a short-term BSc project.)
doi.org/10.1016/j.humo…
Criticism: The experimental group did online movement activities for several weeks, but the controls did nothing. The benefits of intervention need narrowing-down in further studies. (This was a short-term BSc project.)
doi.org/10.1016/j.humo…
Paper #42: Blanchard et al. 2017
Criticism: This comes from a (still! ongoing) project with the fabulous Dr Blanchard & (nearly-Dr) @BlandineFrench. This paper feels exploratory, written before the project ended; almost a 'practice' paper. Is that OK?
doi.org/10.3389/fpsyg.…
Criticism: This comes from a (still! ongoing) project with the fabulous Dr Blanchard & (nearly-Dr) @BlandineFrench. This paper feels exploratory, written before the project ended; almost a 'practice' paper. Is that OK?
doi.org/10.3389/fpsyg.…
Paper #43: Reader et al. 2018
Criticism: Shortly after publication, we were reviewed on @PubPeer. It hurt, but I think we agree with some points. Like the criticism of Mozart in Amadeus, there are "too many p-values" in this paper :-) @ArranReader
doi.org/10.1111/ejn.13…
Criticism: Shortly after publication, we were reviewed on @PubPeer. It hurt, but I think we agree with some points. Like the criticism of Mozart in Amadeus, there are "too many p-values" in this paper :-) @ArranReader
doi.org/10.1111/ejn.13…
Papers #44-57 will come tomorrow, or later still...
I was going to take Good Friday off work like any sensible human should. But certain world events have upset my weekend plans, so I'm dining alone tonight :-(
This has been strangely fun, interesting, & difficult...
I was going to take Good Friday off work like any sensible human should. But certain world events have upset my weekend plans, so I'm dining alone tonight :-(
This has been strangely fun, interesting, & difficult...
I have a compelling, almost divine urge to get this thread done before Easter Sunday, so here we go!
#SelfCriticism
#SelfCriticism
Paper #44: Holmes & Meteyard 2018 @abcdlabReading
Criticism: We argue for a continuous relationship between TMS location, annoyance & effect on behaviour, but my search strategy was discrete: papers from the most- & least-annoying scalp area. Incomplete
doi.org/10.3389/fpsyg.…
Criticism: We argue for a continuous relationship between TMS location, annoyance & effect on behaviour, but my search strategy was discrete: papers from the most- & least-annoying scalp area. Incomplete
doi.org/10.3389/fpsyg.…
Paper #45: Holmes & Tamè
Criticism: Too English. Working in Israel I was impressed by their academic freedom & rigour. Sometimes this caused fights, but was overall positive. My commentary here is actually a harsh criticism of the paper. Can you tell?
doi.org/10.1016/j.cub.…
Criticism: Too English. Working in Israel I was impressed by their academic freedom & rigour. Sometimes this caused fights, but was overall positive. My commentary here is actually a harsh criticism of the paper. Can you tell?
doi.org/10.1016/j.cub.…
Paper #46: French et al. 2018 @BlandineFrench
Criticism: After this critique of the MABC2 test, we heard the MABC3 was in production. I hope our work still influences MABC3, but we could/should have tried to collaborate more & earlier with the MABC guys
doi.org/10.1371/journa…
Criticism: After this critique of the MABC2 test, we heard the MABC3 was in production. I hope our work still influences MABC3, but we could/should have tried to collaborate more & earlier with the MABC guys
doi.org/10.1371/journa…
Paper #47: Reader et al. 2018
Criticism: Part of @ArranReader's 'annus mirabilis' of papers, the main problem here was that we had to discard ~8% of data due to 'false starts' caused by a sub-optimal experimental design and/or data acquisition protocol
doi.org/10.3758/s13414…
Criticism: Part of @ArranReader's 'annus mirabilis' of papers, the main problem here was that we had to discard ~8% of data due to 'false starts' caused by a sub-optimal experimental design and/or data acquisition protocol
doi.org/10.3758/s13414…
Paper #48: McDowell et al. 2018
Criticism: I've made much noise about having good control conditions in #TMS studies. This paper by Tomas devised some truly ingenious methods & analysis that deeply impressed me. But the No-TMS control condition is weak.
doi.org/10.1016/j.cort…
Criticism: I've made much noise about having good control conditions in #TMS studies. This paper by Tomas devised some truly ingenious methods & analysis that deeply impressed me. But the No-TMS control condition is weak.
doi.org/10.1016/j.cort…
Paper #49: Reader & Holmes 2018
Criticism: This expt *should* have worked better. TMS is wrongly labelled a 'virtual lesion' method, but even rTMS over premotor cortex *during movement* can not work so well. We should have done more with the EMG data IMO
doi.org/10.1098/rsos.1…
Criticism: This expt *should* have worked better. TMS is wrongly labelled a 'virtual lesion' method, but even rTMS over premotor cortex *during movement* can not work so well. We should have done more with the EMG data IMO
doi.org/10.1098/rsos.1…
Paper #50: Lakens et al. 2018 @lakens
Criticism: I've been hanging on the citation coat-tails of this crowd-sourced paper. My contribution was mostly replacing '<' with '≤' in the Google document (I joke you not), but I heartily-agree with the message
doi.org/10.1038/s41562…
Criticism: I've been hanging on the citation coat-tails of this crowd-sourced paper. My contribution was mostly replacing '<' with '≤' in the Google document (I joke you not), but I heartily-agree with the message
doi.org/10.1038/s41562…
Paper #51: Reader & Holmes 2018 (what a year!)
Criticism: The most interesting comparisons in this paper are between conditions (meaningful vs. meaningless actions), but the TMS control condition (vertex) is weak. I've long known that vertex is ~ no TMS
doi.org/10.1007/s10548…
Criticism: The most interesting comparisons in this paper are between conditions (meaningful vs. meaningless actions), but the TMS control condition (vertex) is weak. I've long known that vertex is ~ no TMS
doi.org/10.1007/s10548…
Paper #52: Zeni & Holmes 2018
Criticism: I feel this paper isn't finished. Dr Silvia Zeni has more coding skills in her little finger than the rest of @TheHandLab has in all its hands, but PhDs are short & constrained in many other ways. We moved on.
doi.org/10.3389/fpsyg.…
Criticism: I feel this paper isn't finished. Dr Silvia Zeni has more coding skills in her little finger than the rest of @TheHandLab has in all its hands, but PhDs are short & constrained in many other ways. We moved on.
doi.org/10.3389/fpsyg.…
Paper #53: Meteyard & Holmes 2018
Criticism: With @abcdlabReading, planned & funded this small, 1-experiment project. It should have been 100 subjects with both hemispheres & an MRI scan each. Instead it was N=20, one hemisphere, no scan. Needed more £££
doi.org/10.1016/j.jneu…
Criticism: With @abcdlabReading, planned & funded this small, 1-experiment project. It should have been 100 subjects with both hemispheres & an MRI scan each. Instead it was N=20, one hemisphere, no scan. Needed more £££
doi.org/10.1016/j.jneu…
Paper #54: Daly et al. 2018
Criticism: As a control freak, it is hard to let others into the lab to play with my #TMS toys. Dr Daly is a great engineer, but I don't understand the maths or analysis he used in this paper. I just have to trust that it's OK
doi.org/10.1088/1741-2…
Criticism: As a control freak, it is hard to let others into the lab to play with my #TMS toys. Dr Daly is a great engineer, but I don't understand the maths or analysis he used in this paper. I just have to trust that it's OK
doi.org/10.1088/1741-2…
Paper #55: Holmes & Tamè 2019
Criticism: A one-sentence contribution: "S1-hand is 2cm lateral to M1-hand" took me 2 papers & most of a year's work to make. This 11-page review & analysis is too long. I wanted to overwhelm the reader with evidence. I did
doi.org/10.1152/jn.006…
Criticism: A one-sentence contribution: "S1-hand is 2cm lateral to M1-hand" took me 2 papers & most of a year's work to make. This 11-page review & analysis is too long. I wanted to overwhelm the reader with evidence. I did
doi.org/10.1152/jn.006…
Paper 56: Holmes et al. 2019
Criticism: This is my best work, with @TameLuigi & many others. I know it's improvable, but it's too early to see more problems. E1 was underpowered & messy (BSc project!); the fMRI data were not strong or high-res enough.
doi.org/10.1152/jn.006…
Criticism: This is my best work, with @TameLuigi & many others. I know it's improvable, but it's too early to see more problems. E1 was underpowered & messy (BSc project!); the fMRI data were not strong or high-res enough.
doi.org/10.1152/jn.006…
Paper #57: Holmes et al. 2020 (last one!)
Criticism: A 5-expt (failure of) replication, we over-relied on *reported* methods & didn't get the expt working before testing the hypothesis. This is more a running commentary than the strongest possible test.
doi.org/10.1007/s00221…
Criticism: A 5-expt (failure of) replication, we over-relied on *reported* methods & didn't get the expt working before testing the hypothesis. This is more a running commentary than the strongest possible test.
doi.org/10.1007/s00221…
Now that I am free of the worst of my own scientific failings, I will feel a tiny-bit less guilty criticising others' work on my new podcast @BarError #errorbarpodcast
I do try to be fair, but criticism can still be painful. Peace!
I do try to be fair, but criticism can still be painful. Peace!
• • •
Missing some Tweet in this thread? You can try to
force a refresh
