Paddy Maher

New preprint! We're all wrong about who's lonely – and about what loneliness is. By failing to just be direct when we ask people if they're lonely, we've been worried about the wrong people🧵🧵🧵

@binarybits @DKThomp Completely my mistake – the randomisation was per-task. You’re right! The linked material doesn’t make it clear, but the full paper does. Interesting!!

@PaddyCM @DKThomp Did they randomly assign some people to do all AI and some to do no AI? That wouldnt' make much sense.

I'm not sure if there's an AI bubble, but right now I think the "no bubble" side has stronger arguments. People are betting that AI revenues are going to go up a lot over the next couple of years and that seems plausible to me.

@binarybits @DKThomp Yeah, I thought about that too – but while the 246 tasks allow us to be sure that the non-AI group really were ‘better’ than the AI group, they contribute nothing at all to our confidence that this reflects the effect of AI (vs a random skill or speed imbalance across the groups)

@PaddyCM @DKThomp Arguably the n is 246 rather than 16. But also yes IMO the arguments for "strong" AI skepticism are not very good.

@binarybits @DKThomp Great piece, but citing an N = 16 experiment as one of the ‘strongest bits of evidence’ for AI skepticism damages that argument pretty badly

I collaborated with @DKThomp to explain the strongest arguments for and against an AI bubble. Read it here. (The "against" arguments are for subscribers only.) understandingai.org/p/six-reasons-…

@AndyMasley Can’t critique the study, but it’s p counterintuitive given that the narrative over the past few years has been about how genuinely difficult and slow it is to increase supply. Would love some reporting on what’s actually happening here (if this is true)

According to a recent study from the Lawrence Berkeley National Laboratory, data centers seem to have reduced household electricity costs where they're built. I don't think this is the final word, but I'd be interested in any concrete criticism of this study specifically. Like with water, you should be very skeptical of people very confidently claiming they're driving electricity prices up if they can't back that up. #bib40:~:text=Contrary%20to%20these%20concerns%2C%20our%20analysis%20finds%20that%20state-level%20load%20growth%20in%20recent%20years%20(through%202024)%20has%20tended%20to%20reduce%20average%20retail%20electricity%20prices" target="_blank" rel="nofollow noopener">sciencedirect.com/science/articl….

@jbarro Not to defend anyone involved, but I think you’re skipping two points: 1 - an incumbent has to campaign and be president at the same time, and 2 - 100 hours/week of travel & yelling to crowds is probably harder than 100 hours/week of internal meetings and sleeping in your own bed

The argument Harris makes here -- KJP has made it, too -- is that Biden was well enough to *be* president even if he wasn't well enough to run for president. The only sense in which that's true is you can outsource official functions to staff in a way you can't with campaigning.

@dr_appie Amazing piece of work. Really fascinated by loneliness here - how common is it in your experience that a phenotypic r will be positive while the genotypic r is negative?

Genes linked to sexlessness overlap with genes associated with: - Higher education & IQ - Less substance use - Higher autism & anorexia risk - Lower ADHD, anxiety, depression & PTSD risk

The largest study on late life virginity, based on > 400k individuals, out now in @PNASNews Open access link: pnas.org/doi/10.1073/pn… Shoutout to shared first author @laurawesseldijk ❤️ Thread below 👇🏽

"People that are working here shouldn't be targeted like that, nobody should be targeted like that." Social media content creator and “proud Englishman” John Fisher, aka Big John, on the flag debate. #Newsnight

@Tactical_Times Carra said he heard the VAR feed and that they genuinely deemed it accidental?

A case of VAR not being able to intervene for yellow card offences. I think the referee simply didn’t see it. He wasn’t close enough to goal for it to be considered DOGSO but it was a definite handball.

@__inshorts @s8mb Always bothers me that goodreads can’t or won’t bother to create a basic algorithm and I start to wonder about this

@s8mb have you considered learning to code in Python? could be a useful career pivot

Considering how much Amazon knows about my and other people's book buying and reading, it's really surprising how bad its book recommendations are.

@jburnmurdoch Fascinating, ‘easily distracted’ might easily be fixed ability in an increasingly attention-competitive environment, but that wouldn’t explain the drop off in following through/perseverance. Persuasive stuff here

If the headline terms still feel fuzzy, we can dig into the more detailed traits they’re made up of. Here are some of the sub-traits inside conscientiousness: Young people say they increasingly struggle to make plans and follow through on them. They feel distracted, careless.

NEW: Is the internet changing our personalities for the worse? Conscientiousness and extroversion are down, neuroticism up, with young adults leading the charge. This is a really consequential shift, and there’s a lot going on here, so let’s get into the weeds 🧵

There are two common misconceptions in psychology that go hand in hand: 1) that you need loads of (usually repetitive) items to measure the same construct and 2) that you cannot use direct language when asking about that construct. We argue both of them below 🙂‍↕️👇

Huge huge thanks to @YDragostinov, @RStewartPsy, @renemottus, my supervisor @DrAndyCooper, and @tuckerdrob for working on this with me. Loads (loads) more to come osf.io/preprints/psya…

By beating around the bush, we've been systematically exaggerating the link between loneliness and social support ➡️ we've been misunderstanding where lonely feelings come from ➡️ we've been misunderstanding what to do about loneliness Just ask people if they feel lonely!

New preprint! We're all wrong about who's lonely – and about what loneliness is. By failing to just be direct when we ask people if they're lonely, we've been worried about the wrong people🧵🧵🧵