Most advice about evening light is written as if there were one safe brightness for everyone after dark. The research suggests there is not. I think that changes the question. It is less "is my lounge room too bright?" and more "how bright is too bright for me?", which nobody can answer by looking at a light bulb. In this clip, Prof Sean Cain, a sleep researcher at Flinders University and senior author of a 2019 lab study on how people respond to evening light, explains how far apart people turned out to be. The same team then measured the light in 59 homes over the three hours before bed. Nearly half were bright enough to halve melatonin, the hormone that tells the body it is night, and the average home was predicted to suppress it by anywhere from 0 to 87%, depending on whose sensitivity was applied. Energy-efficient bulbs gave off nearly twice the kind of light the body clock responds to as old incandescent ones. Both studies were small, the lab study used healthy young adults, and melatonin is a marker rather than a measure of sleep or health. So it shows how differently bodies respond, not who will sleep worse. Most of us will never be tested, so the sensible move is the free one: turn the main lights down for the last few hours before bed. The same lamp is not the same evening for everyone.

Two weeks with no internet on their phones, and 91% of people in one trial improved on at least one of the three things the researchers measured: wellbeing, mental health and attention. The trial, led by Dr Noah Castelo at the University of Alberta, randomly assigned 467 people to block all mobile internet on their phones, either straight away or two weeks later. Calls and texts still worked. In the group that blocked first, screen time fell from 314 to 161 minutes a day, and on a standard attention test the average gain was about the size of 10 years of age-related decline. That is one task and a small average change, not a reversal of ageing. I think that makes it worth a two-week try, because it costs nothing and you will know fairly quickly whether it helps you. In the clip, Castelo explains what happened when people slipped, which is the part that makes it realistic: • Block the internet on your phone with a blocker app, keep calls and texts, and save online jobs for a laptop. • Charge the phone in another room overnight, so it is not the first thing you reach for. • Decide ahead of time what fills the gap, such as a walk, a book or ringing someone. • Check in after two weeks and keep whatever helped. The people in the trial wanted to cut back, and the comparison group knew they were only waiting their turn, so the results may not hold for everyone. Still, two weeks is a cheap way to find out whether they hold for you.

Taking an autistic child outside is an easy idea to sell: trees, fresh air, room to run. However, a bush clearing can feel just as hard as a classroom if the adults bring the classroom's rules out with them. I think the place and the people are two separate questions, and we tend to answer only the first. In this clip, Dr Samantha Friedman, a lecturer in applied psychology at the University of Edinburgh, describes what she took from her time at a forest school, that is, regular outdoor sessions in woodland, run at a specialist autism school in England. The setting still counts for something, and a separate study shows one way it does. Mateer and colleagues followed 12 autistic boys aged 11 to 20 at a residential school in New York. Each did morning activity for two weeks on woodland trails and two weeks in a fully equipped indoor gym. On the trails, the boys spent about 27 percentage points more of each session moving briskly, meaning hard enough to raise their heart rate and breathing. That study was small, it came from one school, and it measured how much the boys moved, nothing more, so it cannot speak for every child. So the outdoors may help a young person move more, but how much good it does still depends on how the adults there treat them.

Where do you feel a kettlebell swing the next morning? For a lot of people starting out, the honest answer is the forearms, the shoulders or the lower back. I think that soreness is worth paying attention to, because, as this short explainer of mine shows, the work is meant to land mostly behind you, in the backside and the backs of the thighs. It also helps to know where the one number in the video comes from. It is a small lab study of fit young men, most of them used to kettlebells, doing one-arm swings. In that same study the front abs did very little in a normal swing, and other studies, using different bells and methods, put the hamstrings level with the glutes, which is why the ranking at the end is a guide rather than a league table. So if your arms are hauling the bell up, or your lower back tires before your glutes do, it is worth going lighter until the effort moves back to your hips. Judge a swing by where you feel it, not by how heavy the bell is.

A drought usually gets blamed on the sky. However, part of how hard a dry year bites is decided on the ground, by whether the rain that does fall can soak in and stay there. I think this is the half of a dry year we can actually do something about, because the sky is out of our hands, and the paddock under it is not. Two plain definitions help here. Infiltration is how fast rain soaks into the ground instead of running off it. Soil carbon is what is left of roots, microbes and plants in the soil, and it is what makes good soil dark and crumbly. In this clip, former CSIRO soil microbiologist Walter Jehne pushes a spade into a paddock at Langhorne Creek in South Australia, in the middle of a February summer, and shows what soil that can drink looks like: • The spade goes in easily, which he says means an open, loose structure. • Roots run right through it, and the soil is dark, high in carbon and full of aggregates, meaning crumbs that hold together when wet. • That "soil carbon sponge", in his words, is critical whatever the rainfall is, for getting the water in and storing it. • His task for "all over Australia" is rebuilding how well our soils take in and hold the rain. The research backs the first half of that more firmly than the second. A 2019 review of field trials found that ground kept under living roots and cover let rain in faster than bare, worked ground. More carbon does help soil hold water that plants can use, but by less than the figures shared online suggest, and it depends a great deal on the type of soil; one large review of the evidence called the effect limited. So the bigger difference is often whether the rain gets in at all. A crusted, bare paddock can shed a storm that a covered, crumbly one would drink. None of this means soil can end a drought, but how dry a year gets on the ground depends partly on whether the soil can drink.

Your blood test comes back, the fasting glucose line sits comfortably in the normal range, and it feels like one less thing to think about. However, a normal glucose result tells you blood sugar is being held steady, not how hard the body is working to hold it there. I think this is the part of blood sugar most people never hear about, because the number that usually gets checked is one of the last to move. Insulin resistance means the body's cells respond less to insulin, the hormone that moves sugar out of the blood, so the body makes more of it to keep blood sugar steady. In this clip from a 2018 BYU forum address, Ben Bikman, a BYU professor of cell biology and physiology, explains how that extra effort can go unnoticed. The timescale he gives is his own argument. A long-running study of British civil servants, known as Whitehall II, gives some outside support. Among those who went on to develop type 2 diabetes, fasting glucose stayed in the normal range for years and only climbed steeply in the last few years before diagnosis, yet their insulin sensitivity was already lower more than a decade earlier. It is an observational study of group averages, mostly in white British men, so it shows a pattern rather than any one person's path. Most people with a normal result are simply fine. It is just worth knowing what the number can show: that the body is coping, not whether it is at ease.

The glass of orange juice at breakfast often gets counted as a serving of fruit. However, juice is not fruit with the inconvenient bits removed, because the bits that get removed are what change how the sugar lands. I think it's one of the easier swaps at breakfast: eat the orange, and have water with it. In this clip, Dr Robert Lustig, an emeritus professor of paediatrics at UCSF whose work focuses on childhood obesity and diabetes, explains why the fruit and the glass behave so differently. One refinement to his wording: fibre doesn't stop sugar being absorbed, it slows it down. In a small 1977 study of healthy adults given whole apples, apple purée or apple juice, the juice went down much faster, left people less full, and was followed by a sharper drop in blood sugar afterwards. None of this makes juice a poison, or a reason to worry about what's in a child's lunchbox. It's the same sugar, but it doesn't arrive the same way.

At ten at night the phone switches to its warm amber tint, and it feels as though the evening is now sorted. The overhead lights in the lounge, meanwhile, stay on full until bed. I think we have been fixing the smaller of the two lights. If brightness is what the body clock mostly responds to, then dimming the room after dinner is likely to do more than changing the colour of a screen. Prof Russell Foster, who leads the Sleep and Circadian Neuroscience Institute at Oxford, makes the case in this clip. He is one of the scientists who discovered the light sensors in the eye that help set the body clock, so few people are better placed to say what those sensors respond to. The trial evidence on night mode points the same way. In a 2021 study, 167 university students were randomly assigned to an hour of phone use before bed with Apple's Night Shift on, the same with it off, or no phone at all, every night for a week, and their sleep did not differ between the groups. That was one small trial in young adults, tracked for a week with wrist sensors, so it cannot settle the question for everyone. It also does not show that screens before bed are harmless. So as the evening winds down, it may be worth reaching for the light switch before the phone settings.

You went to the gym this morning, so the next eight hours in a chair can feel like they should not count against you. However, Keith Diaz, an exercise scientist at Columbia University, has spent his career studying the harms of being highly sedentary, meaning sitting or barely moving for long stretches, and in this clip from his TED talk he describes a study from his lab. I think what he found shifts the question from "did I work out today?" to "how long did I stay still?", which is a much easier one to act on, because it asks for so little at a time. It helps to know how the study was set up. Eleven middle-aged and older adults without chronic conditions spent eight-hour days sitting in the lab, once with no walking at all, and on other days with short walking breaks of different lengths and frequencies. The team then tracked their blood sugar, the amount of sugar in the blood, and their blood pressure through the day. Keep the limits in mind. It was a small group, each day was a single lab session with set meals, and it measured blood sugar and blood pressure on the day, not anyone's health over the years. A walk is not a replacement for treatment from a doctor. So for anyone who sits most of the day, the interesting part of his answer is not the intensity, but how often.

Picture someone in their 50s or 60s who has never trained a day in their life. Most of us would guess the window for building muscle has closed, and that a few weeks of lifting could not change much. I think that guess sells people short. Muscle does respond more slowly as we age, but lifting something heavy still seems to pay off well into old age. Brad Schoenfeld is a professor of exercise science at Lehman College in New York, and on The Proof with Simon Hill podcast he was asked this exact question. In this clip he describes a meta-analysis his group ran, meaning they pooled the results of many randomised trials, where people are assigned by chance to either train or not. A "standard deviation", the unit he uses for the size of the effects, is just a standard yardstick for how big the gap is between the two groups, and he reports strength and muscle size separately. Keep the limits in mind. The people in those trials were very elderly, mostly training under supervision, and the trials measured strength and muscle size, not how long anyone lived or how independent they stayed. So for anyone starting late, the better question may not be whether their muscles can still respond, but whether they will give them something heavy to respond to.