Nick Holmes
LightNews LightNews
banner
thehandlab.bsky.social
Nick Holmes
@thehandlab.bsky.social
1.8K followers 940 following 820 posts
handy neuroscientist | touch & movement | brain stimulation | neurobiography.info | pod @theerrorbar.bsky.social theerrorbar.com | @tmsmultilab.bsky.social https://tms-rat.org https://tms-smart.info | b. 335ppm C💚2 | pro birds | Associating with Professors
Posts Replies Media Videos
thehandlab.bsky.social
It's a barn owl, right? 🤩

waterbeachwb.co.uk/wp-content/u...
November 7, 2025 at 12:23 PM
thehandlab.bsky.social
Hello #Bird lovers. What the hell is this big feathery thing flying through my garden at midday in Warwickshire UK?

Owl? Red kite? Buzzard? 🤔

At this position from the camera, this is a big bird!

#Ornithology #Birding #Twitcher
A large bird flying around a tree at midday in Warwickshire, UK. I don't know what kind...
November 7, 2025 at 10:33 AM
thehandlab.bsky.social
Deep.
Notice on train schedule app: "This train has been delayed by this train being late"
November 6, 2025 at 7:40 PM
thehandlab.bsky.social
what fresh hell is this?

(error when viewing a work document in a work email from a work colleague on a work computer)
"Microsoft Word: Sorry, your session will expire at 22:52. You can refresh the page to continue."
November 4, 2025 at 11:00 PM
thehandlab.bsky.social
After about four years of regular hedgehogs 🦔 and the occasional fox 🦊 appearing on our back garden trail camera:

Together at last! #FoxOfTheDay

(No mammals suffered in the making of this image. Hedgehogs are 💪.)
A back garden at night, caught on camera. A curious fox on the left investigates a distracted (maybe sleeping, likely eating) hedgehog on the right.
October 17, 2025 at 6:07 AM
thehandlab.bsky.social
Say what you like about the philosophy of science, but this dude is on fire! 🔥

(Feyerabend 1975/1993, p163)

#MetaScience #ScientificReform #ReplicationCrisis #PhilSci
"Finally, let me repeat that for me the chauvinism of science is a much greater problem than the problem of intellectual pollution. It may even be one of its major causes. Scientists are not content with running their own playpens in accordance with what they regard as the rules of scientific method, they want to universalize these rules, they want them to become part of society at large and they use every means at their disposal - argument, propaganda, pressure tactics, intimidation, lobbying - to achieve their aims. The Chinese Communists recognized the dangers inherent in this chauvinism and they proceeded to remove it. In the process they restored important parts of the intellectual and emotional heritage of the Chinese people and they also improved the practice of medicine. It would be advantage if other governments followed suit." Feyerabend (1975, 1993, p163, appendix to Chapter 15).
October 15, 2025 at 7:36 AM
thehandlab.bsky.social
Stats quiz! There are 4967 questions covering the data collection period from 1877 to the present day. Let's get started!

Q1: What's wrong with this graph 🤔

For context, these are 10 patients with spinal cord injury tested before & after transcutaneous spinal cord stimulation (TSCS)

#stats
October 14, 2025 at 7:13 AM
thehandlab.bsky.social
Although...
September 10, 2025 at 8:04 PM
thehandlab.bsky.social
BREAKING: Ptolemy (c. 369-282 BC) was a p-hacker.

I'll bring you more as the story unfolds. I fully expect a pile-on at Ptolemy's social media accounts in the next 24-48 hours. Hold tight.

#OpenScience #MetaScience #HPS #QRP #P-hacking #ScienceFraud
Footnote 25 in Chapter 14, p.144 of Feyerabend's (1975/1993) 'Against Method': "25. 'Carefully' has been contested by R.R. Newton, <<The Crime of Claudius Ptolemy>>, Baltimore. Newton shows that many of Ptolemy's 'data' were manufactured to fit his model. For his optics this has been known for a long time."
September 9, 2025 at 6:26 AM
thehandlab.bsky.social
I've been sorting through old notes, papers, files for recycling. They ranged from ~1995 to 2015. It was quite a memory trip!

I found a letter from 2012. I'd worked at the University of Reading for 3 years, had worked like a dog, supervised PhD students, published papers, did everything I could...
A one-page letter from the '(in)Human Resources' department at the University of Reading, UK, dated 16th August 2012. It informed me that in the next year I *must* achieve one major grant success, along with some other things. In the UK, 'major' would probably have meant >£200k funding. Grant applications typically take around 3 months to prepare, there are several deadlines each year, you can't re-submit previous grants, and the chance of success is typically around 10-20%. Research - by us academics - has shown little relationship between the quality of grant applications and their likelihood of being funded. It is a lottery. It makes me sad for my former self that I received this letter 13 years ago. The utter c*nts.
September 3, 2025 at 8:44 PM
thehandlab.bsky.social
fitting non-linear models with 10 free parameters to human physiology #TMS data is VERY satisfying indeed.

are there any handy guides out there about principled ways to constrain those parameters? I guess I should just look at each one in turn and check the range is biologically plausible? Or..?

🙏
a motor-evoked potential (MEP) from a human hand muscle, fit using a 10-parameter model (actually 2 components, each with 5 parameters). with enough parameters you can fit anything, so how in principle should I go about constraining those parameters..? (the fit here is r^2=0.999 - not bad!)
September 3, 2025 at 3:31 PM
thehandlab.bsky.social
It took me all summer to read this book and there was nothing at all about LLMs or AI. Very disappointing, give it a miss.
A front page image of Charles Mackay's (1841) 'Extraordinary Popular Delusions and the Madness of Crowds'. The image depicts a cow-like man blowing bubbles through a pipe while a bunch of credulous male prospectors in 19th century Western dress try to reach and catch the bubbles. This monster of a book (~500pages) charts the history of the madness of crowds, including the South Sea Bubble to Tulip Mania, Witchcraft, Alchemy and the Crusades. Why do humans keep on believing the latest fad, craze, or con? Why do we keep getting fooled? It was published 180 years before the 'AI' delusion of the 2020s.
August 20, 2025 at 7:55 AM
thehandlab.bsky.social
The effect on interference type (sensory-, response-, or general interference) was the main question being asked:

The answer suggests that we can't use MEPs as a proxy measure of the stimulus- & response-interference processes that lead up to a manual response.

We need to TMS another brain area..!
the results of the 'slope' parameter across all 15 participants, both hands and two muscles. all are positive (most are significantly positive), and all are similar (no significant effects of condition, for example). The results show that, once a response has been initiated, the corticospinal system(s) increase their excitation in a constant manner. The response to a congruent stimulus is the same as a to an incongruent one.
August 11, 2025 at 11:12 AM
thehandlab.bsky.social
Corticospinal excitability increased before a motor response (we knew that), but:

- only in the responding hand (there is no 'spill-over' into the non-responding hand)

- not in the non-responding hand (there is no 'inhibition' of the non-responding hand)

- no dependence on interference type

4/n
Example data (truly representative data, not just the 'cleanest' or 'best' one) of how the size of motor evoked potentials (MEPs) was modeled in the study. The x-axis is the time before the RT that a TMS pulse was presented. 0ms on the x-axis corresponds to when the response was detected in the grip force measurement. The y-axis is the (normalised to a range of 0 to 1) MEP amplitude - a proxy measure of how excitable the corticospinal tract was. The typical result is illustrated - up to about 200ms before the RT, no change in MEP amplitude is seen (the flat baseline period). At around 200ms before the RT, MEP amplitude starts to increase (the decision time), and this increase is approximately linear (the slope). None of this is surprising - the motor system activates when the motor system activates. The most interesting part is that we can detect changes in the motor system starting about 150ms before the eventual reaction time (which was usually 400-500ms). This is neat! It suggests that some parts of the brain know which is the correct response to make quite a long time before that response is made. And the other neat bit is that the decision time and slope parameters did not change as a function of the experimental condition (whether stimulus- or response-conflict). This suggests that the corticospinal activity is not measuring the interference at all, and is just indexing the 'downstream' processing once a decision has been made.
August 11, 2025 at 11:12 AM
thehandlab.bsky.social
I'm happy to report that @valedichiaro.bsky.social 's 2nd PhD paper is in press! :-). Brief explainer...

- 2-coil #TMS over left & right primary motor cortex
- TMS in the ~300ms before one hand responds
- corticospinal excitability increases before response
- no changes in non-responding hand

1/n
image describing the trials in the experiment. at the start of each trial, a random 1-3 foreperiod contains a fixation cross to warn the participant. The first event in the trial is a 200ms period with two coloured squares left and right of the (now disappeared) central fixation point (both are red, blue, yellow or green). These are the distractor 'Flanker' stimuli. A third coloured square is then presented centrally. This is the target stimulus. If the target is red or blue, the participant must grip a left force meter; if green or yellow, they grip the right force meter. They try to respond as quickly and as correctly as possible. In the ~300ms before their response, two pulses of TMS are delivered, one over each motor cortex. This TMS pulse excites the corticospinal tract and results in a muscle contraction. The size of the muscle contraction is used as a proxy for corticospinal excitability and response preparation.
August 11, 2025 at 11:12 AM
thehandlab.bsky.social
I expected 'Against method' to be a difficult read, but after Feyerabend's Preface, I think it's going to be quite an adventure! #HPS #anarchy 🧪
"Anger at the wanton destruction of cultural achievements from which we all could have learned, at the conceited assurance with which some intellectuals interfere with the lives of people, and contempt for the treacly phrases they use to embellish their misdeeds was and still is the motive force behind my work." Feyerabend (1975, 1993, 3rd ed. p8)
August 5, 2025 at 7:20 AM
thehandlab.bsky.social
If these delicious, exciting cucumbers do not win first prize in my village show today, well then I'll have to move house.
August 2, 2025 at 10:19 AM
thehandlab.bsky.social
Avocet in Kent
Smartphone + binoculars 😄
An avocet stands in perfectly still, shallow water so it's reflection appears directly below its legs.
July 24, 2025 at 6:26 AM
thehandlab.bsky.social
Its not clear who (many) university websites are for. Searching for 'staff list' on this rebranded UK website yields a 12 year old article about psychological resilience as the top result (not clear why it defaulted to search the research archive rather than, eg, lists of its staff) 🤷‍♂️
July 23, 2025 at 10:07 AM
thehandlab.bsky.social
We have the same ones 😊. Here is a very friendly one in Stirling University, Scotland
July 19, 2025 at 10:11 PM
thehandlab.bsky.social
Just fuck already, this is tiresome.
"Trump threatens to set Doge on Musk as pair feud again over budget plan". BBC Headline 1st July 2025
July 1, 2025 at 10:37 PM
thehandlab.bsky.social
“They said ‘he will start a war’. I’m not going to start a war, I’m going to stop the wars.”

I guess #Trump was talking about foreign wars #LA
Trump sends another 2000 National Guards and 700 Marines to LA on fourth day of unrest
June 10, 2025 at 4:45 AM
thehandlab.bsky.social
i.pinimg.com/originals/72...
June 7, 2025 at 1:52 PM
thehandlab.bsky.social
Yes, this!

I once converted (re-anakysed from reported effects) a whole phd thesis' Bayesian stats into t-tests (fun day...). It was very useful for detecting errors, but otherwise, same-same...
Two graphs showing Bayesian stats on a logarithmic x-axis against the equivalent t or F stats on a linear y-axis. Two very straight lines of data are shown. Outliers and anomalies were due only to typos, rounding, or floor effects (not enough decimals!)
June 4, 2025 at 3:57 AM
thehandlab.bsky.social
"There is also a self-righteous aspect to whistle-blowing, so that some people - even whole university departments or academic sub-fields - seem to relish challenging the integrity of the procedures and conclusions of others."

Abelson (1995/2009, p103) on #p-hacking ("fishiness") and #metascience
A one page quotation, including the skeeted text
May 29, 2025 at 7:12 AM