iiinova — where your digital self keeps evolving

Mark Haselgrove @markhaselgrove.bsky.social · 10h

24 month FT Research Assistant role with the excellent Dr Victoria Newell on a project to refine, validate, and implement the Self-Harm Questionnaire – Autism (SHQ-A), a new assessment tool co-developed with autistic people. Details here... jobs.nottingham.ac.uk/vacancy.aspx...

jobs.nottingham.ac.uk

Job Vacancy at the University of Nottingham: Research Assistant (Fixed Term)

Applications are invited for a Research Assistant to work with Dr Victoria Newell on a Medical Research Foundation-funded project to refine, validate, and implement the Self-Harm Questionnaire – Autis...

011

Mark Haselgrove @markhaselgrove.bsky.social · 19h

Great to see the boom in this announcement.

000

Mark Haselgrove @markhaselgrove.bsky.social · 05/10/2026

Great hypothesis. Nightmare ethics application.

040

Mark Haselgrove @markhaselgrove.bsky.social · 04/10/2026

Sunday morning Ethel bagging in the Peak District. Lots of geological sights, a good trig point, and then a pint and a packet. Wardlow Hay Cop, high above Cressbrook Dale. Ethel number 37/95.

030

Reposted by Mark Haselgrove

Jialin Shi @jialinshi.bsky.social · 03/10/2026

Check out our W-ACE model: how we use Rescorla-Wagner to explain standard/reverse latent inhibition!!

074

Reposted by Mark Haselgrove

Mark Haselgrove @markhaselgrove.bsky.social · 28/09/2026

New Preprint! "Facilitation and Attenuation of Learning to Novel and Familiar Stimuli Through Contextual Mismatch: Experimental Evidence and a Computational Model" By @jialinshi.bsky.social and me. osf.io/preprints/ps...

osf.io

OSF

1129

Mark Haselgrove @markhaselgrove.bsky.social · 28/09/2026

Honestly APA Editorial Manager, why does this take sooooooooo long. I folded two piles of washing in the time it took you build it

031

Mark Haselgrove @markhaselgrove.bsky.social · 28/09/2026

We regard this as a "proof of concept". The goal is to show that the effect of familiarisation doesn't have to be the loss of salience (e.g Pearce & Hall, 1980 etc) but instead the replacement of one feature with another. With this, a simple delta-rule based model of learning may have some merit.

010

Mark Haselgrove @markhaselgrove.bsky.social · 28/09/2026

Using a simple delta rule (Rescorla-Wagner) tweaked so that the associative strength of the common elements is weighted by its exposure-dependent activation we can simulate our results, showing latent inhibition, and its reversal:

101

Mark Haselgrove @markhaselgrove.bsky.social · 28/09/2026

To explain this, we propose that novelty and familiarity are common elements, permitting generalization between context and cues. As stimuli are exposed, rather than become less salient overall, activity in one element is replaced with another:

100

Mark Haselgrove @markhaselgrove.bsky.social · 28/09/2026

When the context (operationalsed as distractors) sustains familiarity, then learning is more successful to a novel than a familiar cue - latent inhibition But the opposite is observed when the context sustains novelty, ever changing - reversed latent inhibition (see also: Haselgrove et al., 2025)

100

Mark Haselgrove @markhaselgrove.bsky.social · 28/09/2026

Here we report three experiments using real and artificial images which show that the success of learning about familiar and novel stimuli depends on novelty mismatch with the context.

100

Mark Haselgrove @markhaselgrove.bsky.social · 28/09/2026

osf.io

OSF

1129

Mark Haselgrove @markhaselgrove.bsky.social · 26/09/2026

Thanks. Should be through moderation in a couple of days. It too is, as you say, RW adjacent. A generous pinch of RW 😀

010

Mark Haselgrove @markhaselgrove.bsky.social · 26/09/2026

Teaser... It's not* the Rescorla-Wagner model *It is.

070

Mark Haselgrove @markhaselgrove.bsky.social · 26/09/2026

Quite excited by this one. Coming soon to a pre-print server near you...

230

Mark Haselgrove @markhaselgrove.bsky.social · 24/09/2026

Had the first meeting of the incoming editorial board of the @qjep.bsky.social. Lovely to meet everyone, and share our reflections and thoughts. Looking forward to working with the new editor in chief, and the rest of the team over the next few years.

040

Mark Haselgrove @markhaselgrove.bsky.social · 22/09/2026

Completing a reference request form that is so badly designed, I have to use the "show all" function in MS Word to work out how on earth actually input information into it.

010

Mark Haselgrove @markhaselgrove.bsky.social · 19/09/2026

After weeks of nagging from B&Q for feedback from me after I bought a door from them, I have finally relented...

050

Mark Haselgrove @markhaselgrove.bsky.social · 19/09/2026

I've noticed I'm enjoying, more and more, the switching between these perspectives. I have no idea if it has any kind of specific cognitive benefit, beyond the mental health improvements known about from getting out and doing exercise, but I'm sure it can't hurt.

020

Mark Haselgrove @markhaselgrove.bsky.social · 19/09/2026

Other times you are looking miles away and either hours ahead or behind you on your walk, thinking about the bigger picture of your route, or even just appreciating the view.

A border collie stands on a path upon the side of a landscape. Other moorland, hills, woodland and fields can be seen in the distance

110

Mark Haselgrove @markhaselgrove.bsky.social · 19/09/2026

Sometimes you are looking no further than a few meters ahead, working out where to place your feet so you don't sink in a bog or tumble over a rock.

Flagstones sunk into moorland heading off into the distance, running adjacent is a dry stone wall and heather.

100

Mark Haselgrove @markhaselgrove.bsky.social · 19/09/2026

One thing that I am becoming more attuned to while walking, especially out in the wilder parts, is how much switching there is between proximal, short-term processing and more distal longer-term planning.

130

Mark Haselgrove @markhaselgrove.bsky.social · 18/09/2026

Excellent day trip to York, to meet old mates, talk science, plan writing, drink beer, swap gossip, remember the good old days, and then down a curry.

060

Mark Haselgrove @markhaselgrove.bsky.social · 16/09/2026

I've heard that the Vicarage road end of Piddington want to further split from the village, The Arncott road and Lower End group are just not divisive enough. It's just one guy called Ken and his bulldog.

081

Mark Haselgrove @markhaselgrove.bsky.social · 16/09/2026

In a sufficiently large academic institution, there will be enough people who are dicks in slightly different ways, so that ultimately everyone else does nothing else with their time except attend training courses.

1193

Mark Haselgrove @markhaselgrove.bsky.social · 14/09/2026

I asked Claude for a link to the Widrow-Hoff ADALINE model, which it did, but then added a "fun" fact. Maybe we should be worried about the end of the world!

020

Mark Haselgrove @markhaselgrove.bsky.social · 14/09/2026

"Dynamic salience as a basis for behavioural meta-malleability" by Sosa & Trevino. A synthesis of behavioural, computational and neural work on acquired salience. Particularly like their treatment of the Esber-Haselgrove theory of learning & attention. royalsocietypublishing.org/rsos/article...

royalsocietypublishing.org

Dynamic salience as a basis for behavioural meta-malleability

Abstract. Salience, the degree to which cues capture attention, is not fixed but updated through experience. Learning which cues anticipate meaningful outc

041

Mark Haselgrove @markhaselgrove.bsky.social · 14/09/2026

Look very hard for evidence of crime in area A Don't look for evidence of crime in area B Result: Evidence of more crime in Area A than B. It's quite clear what they are doing.

040

Mark Haselgrove @markhaselgrove.bsky.social · 13/09/2026

Best outcome for this pasta:saucepan ratio

021

Mark Haselgrove @markhaselgrove.bsky.social · 13/09/2026

90 s of guinea pig ASMR for you from Coffee and Toffee (sound on)

010

Reposted by Mark Haselgrove

Aidan Horner @aidanhorner.bsky.social · 11/09/2026

We're hiring! The York Neuroimaging Centre (YNiC) is looking for a Senior Technical Specialist to support & develop MRI. This is a new role to reflect our commitment to pushing forward MRI research at York. Come join the team! jobs.york.ac.uk/vacancy/seni... #neurojobs #neuroskyence #psychscisky

jobs.york.ac.uk

Jobs - The University of York

12428

Mark Haselgrove @markhaselgrove.bsky.social · 12/09/2026

Just started season 1, really enjoying it. Rather like a version of Inspector Morse but which doesn't take itself so seriously.

111

Mark Haselgrove @markhaselgrove.bsky.social · 11/09/2026

It's taken me decades to get the point of avocado, and I'm still not fully sold on it, but it added something (smashed and with lemon juice) to my garlic mushrooms on toast.

120

Mark Haselgrove @markhaselgrove.bsky.social · 11/09/2026

The holidays are coming...The holidays are coming... #Christmas

110

Mark Haselgrove @markhaselgrove.bsky.social · 08/09/2026

Watched "Contact" for the first time in a while and was reminded of "that" shot which is just mind blowing. A great deconstruction of it here... youtu.be/HQRu9cz5L9E?...

youtu.be

Cinematographer deconstructs THAT shot in 'Contact' | Don Burgess

YouTube video by Cooke Optics

001

Mark Haselgrove @markhaselgrove.bsky.social · 08/09/2026

For me, it would be valuable to not fall into his trap. He is trying to brand development funding as wasted tax payers money. It's not. It's called being part of a civilization, and recognising the economic inequality between the global north and south. Appreciate the fact checking though!

060

Reposted by Mark Haselgrove

Mark Haselgrove @markhaselgrove.bsky.social · 04/09/2026

Apart from the financial irregularity in all this, one thing that really struck me was the language used by the pollster who described (at 8 minutes) how he was approached and told "we want to do this polling of councils to show x"

221

Mark Haselgrove @markhaselgrove.bsky.social · 04/09/2026

Agreed. Plastic bags for grown ups.

010

Mark Haselgrove @markhaselgrove.bsky.social · 04/09/2026

When done ahead of a democratic process, this is very dangerous. The conformity bias set up here is clear. "Poll shows most people believe X". The fact that the poll was politically motivated, and by the sounds of it, biased, really sounds alarm bells.

000

Mark Haselgrove @markhaselgrove.bsky.social · 04/09/2026

This shouldn't be how polls work, i.e. set up to SHOW something. They should be free from leading questions and bias as much as possible to REFLECT prevailing attitudes and opinions. Here polls, it seems, have been designed to SHOW something, and then the results published in the media.

020

Mark Haselgrove @markhaselgrove.bsky.social · 04/09/2026

221