john Gill technology header image

Designing an Accessible Membrane Keypad - An Investigation of Tactile Features

Conducted as part of the MediVoice Project (a Co-operative Project, funded by the European Commission under FP6)
by Antoinette Fennell
February 2007

 

The tactile features of membrane keypads were tested with a sample of blind and partially sighted test subjects. The investigation covered button shape, button size, pressure required to press the buttons and button texture.

Eleven blind and partially sighted test subjects (seven men, four women) were asked to test and comment on the tactile features of membrane keypad buttons. The age range of the female participants was 40-79 years (Median = 54.5), and of the male participants was 36-70 years (Median = 58). Three of the test subjects mentioned that they have arthritis.

The membrane keypads consisted of two rows of six buttons. One row had square buttons, the second row had round buttons. From left to right the button size decreased from 12mm to 7mm, in 1mm increments. From left to right the pressure required to press the button increased. Three different keypad textures were tested, categorised as 'glossy', 'matt' and 'rough'. The 'glossy' material was EBG polyester. The 'matt' material was Autotex polyester. The 'rough' material was Lexan polycarbonate.

A sample keypad with two different materials on the one keypad was also tested. The keys on this keypad were glossy EBG polyester and the surrounding area was a rough polyester.

Button Shape

Two membrane keypad button shapes were evaluated: square and round.

All of the test subjects successfully recognised the square shape (McNemar's test: n = 11, p < 0.01). All of the test subjects successfully recognised the round shape (McNemar's test: n = 11, p < 0.01). This suggests that, with regard to tactile navigation of a keypad, it would be acceptable to use button shape as a means of distinguishing between buttons.

There was no significant preference for either square or round (Cochrane's Q = 0.73, n = 11, n.s.).

 

Button Size

Six different button sizes were presented on one sheet, for both button shapes. The button sizes were 12mm, 11mm, 10mm, 9mm, 8mm and 7mm. These measurements represent the length and width of the square buttons and the diameter of the round buttons.

Not all of the test subjects successfully recognised the fact that different sized buttons were present (McNemar's test: n = 11, n.s.) This suggests that, with regard to tactile navigation of a keypad, it would not be acceptable to use button size alone as a means of distinguishing between buttons.

There was a significant degree of variation in preference for the six different square button sizes (Cochrane's Q = 12.2, df = 5, p < 0.05). The 12mm square button size was most popular.

There was a significant degree of variation in preference for the six different round button sizes (Cochrane's Q = 16.1, df = 5, p < 0.01). The 12mm round button size was most popular.

 

Button Pressure

Six different button pressures were presented on one sheet, for both button shapes. The required pressure was a function of button size, type of material used, number of layers of material used, and type of button dome used. As button size decreased from left to right, the pressure required to successfully press the button increased.

All of the test subjects successfully recognised the fact that different degrees of force were required to successfully press each button (McNemar's test: n = 11, p < 0.01).

There was a significant degree of variation in preference for the six different pressure levels (Cochrane's Q = 12.5, df = 5, p < 0.05). The 12mm square button size was most popular, with respect to pressure. This was the button that required the least amount of pressure to press.

There was a significant degree of variation in preference for the six different round button sizes (Cochrane's Q = 13.5, df = 5, p < 0.05). The 12mm round button size was most popular, with respect to pressure. This was the button that required the least amount of pressure to press.

It should be noted that three of the test subjects mentioned that they have arthritis. For these three test subjects, the pressure required to press the 8mm and 7mm buttons was too high and they were unable to successfully press them.

Button Texture

Three different keypad textures were presented to the test subjects: categorised as 'glossy', 'matt' and 'rough'.

All of the test subjects successfully recognised that there were three different textures presented to them (McNemar's test: n = 11, p < 0.01). This suggests that, with regard to tactile navigation of a keypad, it would be acceptable to use button texture as a means of distinguishing between buttons.

When asked for a preference, all of the test subjects chose the 'rough' texture as their favourite of the three (Cochrane's Q = 22.0, df = 2, p < 0.001). This unanimous preference was due to the fact that (1) the buttons on the 'rough' keypad were easier to distinguish, and (2) the surface of the 'rough' keypad was not slippy.

Finally, a keypad with two different textures was presented to the test subjects. This particular sample had a glossy finish on the buttons and a rough finish on the surrounding area (i.e. the rest of the keypad). The test subjects were asked if there was an added benefit to having two different materials on the keypad.

All of the test subjects saw a benefit to having the two different textures on the keypad when compared to the 'glossy' keypad (Binomial Test: n = 11, p < 0.01). All of the test subjects saw a benefit to having the two different textures on the keypad when compared to the 'matt' keypad (Binomial Test: n = 11, p < 0.01). Only three of the eleven test subjects saw a benefit to having the two different textures on the keypad when compared to the 'rough' keypad (Binomial Test: n = 11, n.s.).

Conclusions and Recommendations

In conclusion, the results of this study suggest that, with regard to tactile navigation of a keypad, button shape and button texture are both acceptable means of distinguishing between buttons on a keypad.

With respect to the feel of the button alone (i.e. not counting pressure), in both the square and the round button shape a button size of 12mm was preferred over smaller button sizes. This represents the maximum button size provided in this study.

With respect to button pressure, the lowest pressure on the 12mm button was preferred, with a maximum acceptable pressure of the 9mm button. Above this level test subjects with arthritis were unable to successfully press the buttons.

Regarding button texture, a rough texture provided the best tactile feedback for button distinction. In cases where a rough texture is not feasible, an acceptable alternative is to provide two different textures on the keypad - a smooth texture on the buttons and a rough texture on the surrounding area. It should be noted, that a glossy material can produce glare and can hinder the ease of seeing or reading an icon or label, so a smooth texture without glare would be preferable.

 

Acknowledgements

This project was funded by the European Commission under FP6. The author wishes to thank the anonymous test subjects who took part in this study. Sincere gratitude is extended to the staff of RNIB Bristol, in particular David LePoidevin, for considerable help.

 

 

 



John Gill Technology Limited Footer
John Gill Technology Limited Footer