Two hands is better than one

In everyday life people easily use both hands in complex tasks, like driving a car or doing a portrait. However, when we try to perform activities on a computer, we usually limit ourselves to using our dominant hand for direct manipulations.

Guiard model for two-handed abilities

From breakfast to doing sport, drawing up a list of food products to creating a work of art, we naturally use both hands in everyday life. But, when we use technology we lose this habit. Why?

As human beings we are exceptionally good at coordinating both hands for performing complex tasks, but when we interact with computers we often limit ourselves to using only our dominant hand.

According to the Guiard mode, in two-handed actions, the right hand of right-handed people has an “active” role, while their left hand provides postural support.

Two-handed interaction looks at how to develop systems to allow users to exploit their two-handed interaction ability. This is not only related to daily routine activities but also human-machine interaction.

To successfully use these principles on computers, users must have interfaces available to make two-handed interaction possible. Designers, on the other hand, need to have solid models available to which to refer during the design stage.

Two-handed interactions can be classified as symmetrical / asymmetrical and parallel / serial.

We have symmetrical interactions when both hands are occupied in similar activities; asymmetrical interactions are activities performed by two hands that are different but connected.

Parallel activities are those in which both hands work more or less simultaneously, whereas in serial activities the two hands work in series: the work performed by one hand is input for the activity to be performed by the other.

The first studies on two-handed interaction were based on the assumption that users were sitting at their desks interacting through various input peripherals positioned on their desk.

In October 2016 mobile devices overtook desktops for internet access for the first time.

It is interesting to observe how the use of the various devices changes over the hours of the day.

It should be considered that the data provided relates to the US market and is probably influenced by the type of applications offered by the market.

To date, the availability and popularity of touch screen tablets is increasing dramatically with over 30% of internet users owning one. However, the lack of two-handed interaction in touch devices is providing serious challenges for designers.

Designers sustain that tablets use some intuitive interaction techniques, such as pointing and swiping, but also that they cannot support more critical interaction possibilities. More specifically, tablets are not suitably designed to support two-handed interaction, despite there being sufficient literature to demonstrate that two-handed inputs dramatically increase user satisfaction and performance.

Tablets are a particular challenge for interaction. While small portable devices are designed for one-handed interaction, tablets use software and critical product conventions, extrapolated from computer desktops, such as web browsers.

Conventions, as described by Norman, are learnt restrictions that exclude certain behaviour while promoting other behaviour. For example, convention establishes that all left taps must be cold and all right taps must be hot. Convention also provides for screws to be tightened in the clockwise direction and loosened in the anticlockwise direction.

Conventions do not have to be simply considered as physical constraints. A mouse arrow positioned inside the borders of a monitor is an example of a physical constraint. The user is simply not able to move the arrow with the limits of the screen.

A side sliding bar is an example of a convention, whereas the sliding bar can be limited to the user’s vertical movement. To activate the bar, the user must learn to hold the mouse button down on the bar while moving the mouse vertically. In the same way, even if the physical position and movements of taps are fixed, the user must learn to turn the tap to be able to activate it correctly.

Many of the conventions on current tablets emulate directly from PC conventions that support two-handed interaction. Tablets, on the other hand, limit two-handed interaction, as one hand is immobilized by the need to hold the device still. Therefore, because of the adoption and modification of pre-existing conventions, users are blocked by conventions that were not designed for the interactions they are performing.

Some clear examples are as follows:

  • The size of the buttons and icons of web navigation that were designed for interactions with the mouse without considering the relative size of a finger with respect to a mouse pointer;
  • Users who need to type on virtual keyboards with one hand when they were designed for two-handed typing.

The lack of two-handed interaction in touch screen mobile devices is an area of HCI research that is developing and that has led to the design of various products.

However, to date, no product has been able to live up to the performance of its traditional counterparts.

Originally published at

User Experience e Service Designer. Ha un particolare interesse per l'intelligenza artificiale, le interfacce vocali e il design strategico.