Reading by feel: how to put a drink down so your hand can find it again
How to set a drink down so your hand finds it again mid-chapter — the motor science of the blind reach, and nine fixes that cost nothing.
How to set a drink down so your hand finds it again mid-chapter — the motor science of the blind reach, and nine fixes that cost nothing.
Almost nobody spills a drink while setting it down. Setting down is a movement you watch — both hands honest, the cup in view, the surface in view, at the careful speed of a person carrying something hot. You spill forty minutes later, on the way back, at the bottom of a page you were not willing to leave.
That is a small distinction with a large consequence, because it moves where the fix has to go. If the problem were putting the cup down, the answer would be a better surface. The problem is going back for it — and going back for it is decided in advance, in the two seconds when you set it down, by how much information you left yourself and how findable you made the thing.
What follows is the practical version of that: how a hand actually locates an object it cannot see, why the reach that knocks a mug over is a predictable failure rather than clumsiness, and what to change about where you park a drink so your hand meets it instead of clearing it.
The short version, if the chapter is calling:
You always know where your own hand is. That is proprioception — the sense of self-movement, force and body position, carried by muscle spindles, Golgi tendon organs and joint receptors, and it is the reason a person can walk across a dark room or touch-type without watching their fingers (Wikipedia, Proprioception). Your hand is part of you, so it is never lost.
The mug is not part of you. It has no receptors reporting home. Everything your arm knows about it is a stored coordinate — a snapshot from the last time you looked, held in a memory that has to survive an entire chapter and every time you shifted under the blanket.
Two things therefore have to be true for a blind reach to work: the stored location has to still be roughly right, and the hand has to arrive shaped correctly. Both of them degrade the instant you stop looking, and they degrade in ways that are worth knowing about, because each one points at a different fix.
This is the mechanism behind most reading-chair spills, and it is not intuitive.
When you reach for something under visual uncertainty, your hand opens wider than it needs to. Keefe, Suray and Watt tested this directly in Experimental Brain Research, degrading vision with blur and monocular viewing and measuring how far the hand opened on approach. In ordinary grasping, uncertainty produced significantly larger maximum grip apertures — most strongly for the smallest objects, and the paper's broader point is that the hand adjusts in whichever direction lowers the chance of error for the task in front of it. The premise the authors start from is the intuitive one, that "increased uncertainty typically results in wider opening of the hand", and their data, they conclude, "support the intuitive idea that grasping movements employ a true margin-for-error mechanism" — the motor system erring on the side of caution because it knows its own aim is worse than usual (Keefe et al., 2019).
That safety margin is adaptive when you are grabbing a doorknob. It is precisely the wrong thing when you are reaching for an open mug of tea. A hand that opens wider sweeps a wider corridor through the space where the cup is standing. You are not making a delicate approach with a wrong coordinate; you are making a wide approach with a slightly wrong coordinate. The extra opening that is supposed to guarantee you catch the object is the same extra opening that catches its rim on the way past.
The instinct is that the longer the drink sits there unwatched, the worse your aim gets — that the mug drifts in memory over an hour the way a parked car drifts in a car park.
The evidence is more specific than that, and more useful. Camponogara and Volcic had people reach with the right hand to targets they had been holding with the left but could no longer touch, varying the delay before the movement started: two, four or six seconds. Introducing even the shortest delay cost accuracy and precision — but performance did not keep getting worse across that range, particularly when the encoding arm and the reaching arm were in similar postures. Their conclusion is that postural information may help guide the reach once immediate haptic feedback is gone (Camponogara & Volcic, 2025, bioRxiv preprint — worth flagging as not yet peer reviewed).
Read that back into the armchair — carefully, because six seconds in a lab is not an hour in a chair — and it suggests two things. First, within the window actually tested the penalty behaves more like a step than a slope: you pay it the moment you stop attending to the drink, which is roughly one sentence into the page. What a further hour adds on top of that nobody has measured, so treat it as an open question rather than a free pass. Second, and much more actionable — what held the estimate together in that study was posture that stayed the same. So the thing to protect is a body arranged the way it was when you set the cup down. You tucked a foot up. You slid down against the cushion. You rolled toward the lamp. Every one of those quietly moved the origin your stored coordinate was measured from.
The obvious workaround is to flick your eyes over mid-reach without really leaving the page. It works less often than it feels like it does.
When attention is loaded, people miss things that are in plain sight. That is inattentional blindness — the failure to perceive an unexpected stimulus in view because attention is elsewhere — and its most famous demonstration had about half of viewers, counting basketball passes, fail to notice a person in a gorilla suit walk through the shot. Detection falls sharply as cognitive load rises (Wikipedia, Inattentional blindness). A reader in the third act of a book she has been waiting a month to finish is the high-load condition. The half-glance registers something pale, roughly there and hands your arm the same stale coordinate it already had.
So the fix cannot be a look mid-chapter. It has to be built in before the chapter starts.
Fitts's law puts a number on the trade you are making. Movement time is a function of the ratio between the distance to a target and the width of that target: MT = a + b·log₂(2D/W), from Paul Fitts's 1954 paper on the information capacity of the human motor system (Wikipedia, Fitts's law). Small target, long distance, slow careful movement.
The catch is what you are actually unwilling to spend. It is not the arm's time — it is attention. A slow, deliberate, accurate reach requires you to think about the reach, which means leaving the sentence, which is the one thing the whole manoeuvre exists to avoid. So you spend the other currency. You go fast and imprecise, because being imprecise costs nothing until the once-in-forty-times that it costs everything.
That leaves you three levers, and Fitts tells you which ones are real. You cannot make yourself slower — you have already refused. You cannot make the cup wider. You can shorten the distance, and you can enlarge the effective target: the thing your hand has to hit is not the same as the thing you want to pick up.
None of this requires buying anything. It requires deciding, once, where a drink lives when you read, and then making that decision the same way every night.
1. Park it where your relaxed arm already falls. Not where there happens to be a flat spot. Sit exactly as you read — feet up, blanket on, book open — then let your free arm go completely limp over the side. Where your knuckles land is the address. That posture is the one you will reproduce automatically an hour later without thinking about it, and consistent posture is what went with better accuracy in the reaching study once immediate feedback was gone. A spot two inches outside that arc is a spot you have to aim at.
2. Choose the side your free hand is on, and stop switching. A hardcover in the standard trim wants both hands to keep the spine from closing; a paperback frees one thumb, and it is the same thumb every time. Whichever hand is habitually free is the hand that will do every reach for the rest of your reading life. Put the drink on that side permanently, including in rooms where the other side is more convenient. Convenience is not the variable being optimised here; repeatability is.
3. Look at it once, properly, when you put it down. This is the highest-return second in the entire system and almost nobody spends it. Prior visual information measurably improves the touch search that follows: Jeschke and Drewing found that people given valid visual priors selected the right exploratory hand movement earlier, with higher probability, and finished exploring in less time — they conclude that visual prior information "increases the efficiency of haptic explorations by anticipatory planning of appropriate movement schemes" (Jeschke & Drewing, 2025, i-Perception). One deliberate look, at the cup and at what is around it, and your later blind hand arrives already knowing what it is doing rather than improvising on contact.
4. Anchor it to something that will not move relative to it. Body-referenced memory — about a forearm to my right — is destroyed the first time you shift your weight, which is within about ten minutes. Landmark-referenced memory survives. Set the cup against the arm of the chair on the outboard side, against the wall side of a window seat, tight to the base of the lamp. Then the instruction your hand follows is not a distance from your hip; it is slide until you hit the thing, the cup is just before it. A landmark converts an aim into a stop.
5. Get it up to hand height so you never descend onto it. This is the single most underrated geometric point in the whole problem. If the drink is below your relaxed hand, the only approach available is downward — and a descending hand meets the rim first, which is the worst place on a container to apply a sideways force, because a nudge at the top of a tall tumbler has far more leverage about its base edge than the same nudge low down. If the drink is at or slightly above your relaxed hand height, you can only come at it horizontally, and a horizontal hand meets the side, low, where it can hold rather than topple. This is the real reason the floor beside the armchair is the worst address in the house, ahead of every other reason.
6. Come in along a surface, with the backs of your fingers leading. Free air gives your hand no information until impact. A hand sliding along the chair arm or across the cushion is receiving continuous contact data the whole way in, which is how you convert an aim into a search. Lead with the backs of the fingers rather than the palm: the back of the hand cannot grab, so it cannot clip a rim and pull, and it registers the cup's warmth or cold a moment before it touches. Then rotate the hand and take the cup low.
7. Aim below the widest point. Whatever else you do, the target for the hand is the base of the container, not the rim. It is the part that moves least if you are slightly off, it is the part with the least leverage against you, and on a cup with any taper at all it is the part that guides your fingers upward into a hold instead of outward into a shove.
8. Turn the handle to a known hour. If it has a handle, park it pointing the same way every time — three o'clock, say, square to your line of approach. A handle is a landmark that sticks out into the space your hand is already sweeping. It finds you rather than the other way round, and it is far more forgiving to knuckle than a rim is.
9. Know which drink is actually the dangerous one. It is not the full mug. A full mug is heavy and comparatively hard to shift; it is simply catastrophic when it does go. The easiest thing in the room to knock over is the half-empty can or the two-thirds-gone iced coffee — light enough to move on a knuckle, tall enough to have leverage, and in the iced case wearing a film of condensation that lets it skate rather than tip. If you are going to be reaching for something blind for two hours, that is the one to give the best address to.
The floor beside the chair. It fails every rule at once: below hand height so the approach is a descent, outside any reachable arc so retrieving it requires a full postural change, and that postural change is exactly what invalidates the coordinate you stored. It is also out of the lamp's pool of light, so the corrective glance that might have saved you returns nothing. There is no version of the floor that works.
The overloaded nightstand. Here the target is not the cup — it is the gap between the lamp base and the phone. Fitts's law is unambiguous about narrow targets at distance: they demand slow, deliberate movement, which is the one thing an absorbed reader will not supply. And the penalty for missing is not a spill, it is the lamp.
The top of a book. It keeps happening because a closed hardcover is genuinely the flattest, most level surface anywhere near a reading chair — which is the whole tragedy of it. Readers do not need to be told what that costs; most of us can name the book. If a book is being used as a coaster it is because everything else in the room is soft or sloped, and that is a surface problem, not a discipline problem.
A rolled or pillowed chair arm. Worth being blunt about: a rolled arm crowns. There is no flat spot on it at all, so nothing set on it — of any width, of any design — is standing on level support. It can be made better by getting the contact patch as far down the inboard flank as possible, where the curve is shallowest and the arm is stiffest. It cannot be made good. If your chair has a rolled arm, the honest answer is a flat perch at seat height beside it, and the second-best answer is the flattest part of the seat cushion on your free-hand side, against the chair's inside edge as a landmark.
Black tea is brewed at or near boiling, and a mug of it three minutes old is not a comic hazard. The American Burn Association's thresholds for hot water, as summarised in Wikipedia's article on scalding, are short: at 155°F a burn injury occurs in about one second, at 148°F in about two seconds, at 140°F in about five (Wikipedia, Scalding). A hand that arrives palm-down over the open top of a fresh mug, or that tips one into a lap under a blanket, is in that range.
Which is one more argument for the low, side-on approach with the backs of the fingers leading — it cannot enter the cup — and one small piece of practical advice that has nothing to do with any product: the first few minutes of a hot drink are the minutes to keep it in view. Let the fresh mug sit somewhere you can see it while you are still finding your page. It becomes a blind-reach object once it has stopped being able to hurt you.
Everything above is about reducing the chance that your hand meets the cup wrongly. It cannot get that chance to zero, because the premise of the whole exercise is that you are reaching for something you are not looking at. Some contact you did not intend is going to happen. So the last variable worth thinking about is what the container does when it does.
That is the specific thing Steadi is for. It is one piece of Shore 85A TPU — not silicone — with a base 5.2 inches across that is fixed: it does not expand, contract or deform. It stands 2.8 inches tall, and its fins flex to grip the bottom inch of a container.
The mechanism is worth stating precisely, because it is the part most products in this category are vague about. The fixed 5.2-inch base widens the base of support underneath the drink. Because the support is wider, the container's centre of mass has to travel further horizontally before it passes outside that base — which means tipping requires a much larger tilt than it did before. That is all it is doing, and it is enough. The tack of the TPU does the second job: it resists sliding, which is what actually matters for a sweating iced-coffee glass. It resists tipping and it holds steady. It does not right a drink, it does not level anything, and it will not save you from a direct hard knock.
The part that belongs specifically in an article about finding things by feel is quieter, and it is the one Fitts already told us to look for. You cannot make a mug wider. But the thing your hand has to find is no longer the mug. A hand sweeping in low along a cushion meets a 5.2-inch disc, at a known height, well before it meets the cup standing on it — a bigger target, met lower down, with the contact happening at the part of the assembly with the least leverage over the drink. The base becomes a landmark in its own right: the same shape under whatever you are drinking, in the same place every night, which is exactly the repeatable reference that reaching without vision depends on.
The honest limits, since this is an article for people who have already bought three of these things off the internet and been disappointed:
The point of any of this is not that you will never spill again. It is that the drink stops taking up room in your head.
Right now some part of your attention is permanently assigned to the cup — tracking roughly where it is, flinching slightly when the cat lands, budgeting a small ongoing anxiety against the hardcover it is sitting next to. That is attention you would rather be spending on the page, and it is the real cost of an unfindable drink, paid every night, long before any actual spill.
Give it one address. Look at it once. Reach low, along the surface, with the backs of your fingers out front. And then stop thinking about it, which was the whole idea.
Purchases go through steadilabs.com — the maker, directly.