Exposing touch screen sensitivity in the UI

Question

Given the growth in the use of touchscreens, how come there's never an obvious touchscreen 'sensitivity' control on screen?

Should this control be included in the UI?

I'm thinking about kiosk design in particular here.

EDIT: The problem I was seeing today was a 'fleeting' touch triggering a button on the screen. The problem wouldn't have been occurring if it was possible to adjust the screen to require a longer period of 'touch' before the computer would register that the button had triggered.

Answer 1

Sensitivity of an input device means how much interaction needs to be performed to achieve the wanted result. It can be measured in:

• Distance - how far an analog input device needs to be moved to affect the active element on the screen (e.g., mouse or joystick);
• Time - how long an input device needs to transmit a signal to be considered an interaction (e.g., keyboard or touchscreens);
• Force - how much pressure needs to be applied to be considered an interaction (e.g., Wacom tablets).

In addition, the term "touchscreen" refers to several sensing technologies that require various amounts of force and time to trigger a signal. However, most of them are incapable of measuring the force applied. The three most common ones are:

1. Capacitive touch has become the most common group of technologies (there're several similar methods of varying accuracy - even within the same type). These technologies power the overwhelming majority of today's touchscreens. The downsides include the inability to calculate the pressure force & the recognition of the slightest touches because of the inherent properties of capacitance.
2. Resistive screens transmit the signal from the point where the 2 conductive layers make contact (pressure is applied on the outer shell). The force required to make contact varies depending on the device but is generally more than what capacitive screens need. Resistive screens can supposedly calculate the pressure force but I haven't seen any applications or mass-market products taking advantage of it.
3. Infrared is the cheapest way to touch-enable a large screen. In reality, this is a pseudo-touchscreen technology since no real touching of the screen's surface is necessary: an object just needs to obstruct the IR rays at a particular level close to the screen. As a result, it's incapable of reading the pressure force and is far less accurate than capacitive and resistive technologies.

There's also electromagnetic resonance, which isn't considered a touchscreen technology because it requires the special stylus to recognize interactions with the surface. It was used in the original Tablet PCs in early to mid-00's and is currently used in Wacom Cintiq monitors (Wacom holds the patent to it).

EMR is the most advanced technology but it's the most expensive one despite being in the market for over 20 years so it's not widespread. It currently supports up to 2048 pressure levels for the stylus pen. The native drivers do allow some modification of pressure sensitivity and applications built to take advantage of the technology, such as Adobe Photoshop, have the option of completely disabling pressure sensitivity.

---

Now, that we know the measurements of sensitivity and the capabilities of the modern technologies let's see how touch sensitivity can be adjusted.

It's safe to assume that you were referring to capacitive screens since they are the most commonly used in consumer devices nowadays. Since the technology requires very little force to produce a signal, sensitivity should control the time the force is applied to the screen. The problem with this approach is that a tap or a mouse click lasts generally up to 100ms, which is very short time and is likely to match the duration of an accidental contact. As a result, increasing such measure of sensitivity will require users to touch the screen for unnaturally long periods of time. Therefore, the experience flaw is in the technology's inherent properties and not in the accompanying software.

Technology is also to blame for the poor responsiveness in some resistive and infrared screens because both of them require significant physical force to trigger a signal.

Answer 2

Two problems here; it's usually set at the hardware/OS level, and it's often far too complicated to present to the normal user.

Touch screens aren't supposed to have "sensitivity" like mice or joysticks do. It's direct manipulation, you should either be touching X or not. If a first press doesn't register, you press harder; this is annoying but it works. This is only a big problem if you require dragging gestures on a resistive touch screen. Don't.

Sensitivity is generally a problem with resistive touch screens; this is generally a hardware problem; you have to push one layer onto a different layer (like the two "screens" on the touch screen of a Nintendo DS). The layers either touch or they don't, you can't adjust it. It's digital, not analog.

Furthermore, even if you could integrate sensitivity, imagine how annoying it would be to have to change the sensitivity setting every time you got to a touch screen kiosk. Maybe you don't know how. Maybe you click the sensitivity button on accident. What does sensitivity mean anyway? Why doesn't this screen work?

None of this is relevant to users of Point of Sale/ect kiosks. People aren't setting up a video game, they're pressing buttons. If the buttons don't work like buttons, they're broke.

Answer 3

Even if it were technically possible (which admittedly it is in some specific circumstances), it would seem to me to be a mistake to expose this sort of option to the user.

We build touchscreen kiosks using Zytronic projected capacitive touch (PCT) displays. For aesthetic and maintenance reasons we apply a single sheet of 3–6mm glass over both the display and bezel, so the capacitive touch panel itself is as much as 8mm from the user's finger when they touch the screen.

When you're setting up a panel using standard UPDD drivers, you can control a bunch of responsiveness settings; glass thickness, edge gain and touch threshold among them (as well as touch calibration of course). Since no two kiosks (nor two touch panels) are exactly the same, even when built to the same spec, you generally need to calibrate the touch separately for each unit (ideally once the unit is put in its eventual location you should recalibrate to account for changes in height or other unforeseen circumstances). Getting the touch responsiveness just right such that it responds exactly when you expect it requires a lot of playing around with the actual settings.

Once that work has been done, though, the user shouldn't ever need (or even want) to control the touch responsiveness themselves.

Edit: One software thing you can (and probably should) do is to only trigger button presses on touch up (not when a button is pressed). That allows for the familiar interaction of pressing a button, sliding your finger off the button and releasing outside to prevent an action. There are certain driver settings that will prevent that sort of thing (at least in the UPDD console); you need to enable dragging (in order to prevent accidental selection, especially on the web, many kiosks are configured to register a mouse down and a mouse up event instantaneously when a user touches the screen). It probably goes without saying, but your buttons should also have a clear "depressed" state if you implement this.

Answer 4

One non-expert thought--ability to apply adhesive transparent layers to a cellphone screen to reduce sensitivity? Not casually adjustable, but could solve a general problem, like mine (on HTC Android One X).