Dominic Amato February 15, 2015
Observing Differences in Perception based on the Direction of Movement
To observe the differences in responses to an exercise taken from Bailey’s “Human Performance Engineering” and a similar activity based off of it. The activity would serve to shed insight on differences of understanding of interface designs and layouts between cultures and highlight both consistencies and inconsistencies that the designs contained.
Given the exercise, the easiest method to reach out to a large number of people was utilizing social media (Facebook & Twitter) and having participants fill out an online survey. Using google forms the participants were given an image of the handout and a reference image for clockwise and counter-clockwise movement. Each question on the handout had a corresponding area on the survey with choices clockwise and counter-clockwise written out. In addition to the original questions a second section was added utilizing the same type of questions as the handout only with less variability. The question surrounded the same mechanic, turning a knob to move a dial, this time there was no knob pictured and people were supposed to imagine the knobs location. Such an activity seemed easy enough given they had just done a similar one in the first section. Further the dial never changed orientation and the knob moved in a simple pattern from left to top, bottom, and right so as to see if people made consistent responses based on the patterns. Demographic data was also collected but was not mandatory.
The survey was hosted for 3 days with the hopes that a minimum of 30 people would take it. Once the 3 days was up the survey was closed and the data put into a spreadsheet for analysis.
After analyzing the data there was a lot of variability in the first section and less in the second. A total of 33 people took the survey with 16 females and 17 males participating, 28 people were right handed and 5 were left handed. The average age was 29.77 with the median being 26, a low of 21, and a high of 55. Median education of the participants was a Bachelor’s degree, 9% had at least completed high school, 27% had completed some college, 43% had a Bachelor’s degree, and 21% had a graduate degree. 6 People were from outside America though 7 people did not enter in any cultural information.
Responses are handled in Figure 1. Figure 2 shows a reference image for direction that the users were given, and Figure 3 is the dial from the second section. Questions were:
Figure 1: Response Distribution, Blue for Counter-Clockwise and Orange for Clockwise
Figure 2: Directional Reference Image
Figure 3: Dial Reference for Section 2
After analyzing the responses it was hard to isolate any specific cultural stereotypes from the data. In the future larger more diverse sampling may help indicate any consistent direction of movement choices.
It was interesting that three questions received nearly unanimous responses. Question 4 and 5 of the first section were very interesting as they should have been opposite answers though two people entered the same direction to turn the dial both right and left. This may have been due to the way the survey was laid out as the first section had all the questions and pictures above making participants scroll up and down to reference the image and answer the question. Question 1 in the second section was the only one that was universal among all participants. The question had people suppose a knob was to the left of Figure 3 and they were asked to increase the frequency. This could be because most modern car stereos have this layout which cause people to be familiar with it to either increase volume or frequency. Still when given the question of placing the knob on the right the response data was not universal even given the same image and wording, only they were asked to decrease the frequency.
Question 1 and 6 in the first section had the same response percentages except for 2 participants switched their answers. What is most intriguing about this is that given the numbering layout of question 1, question 6 should have been universally counter-clockwise if people assumed the numbering layout to be the same between them. Again it may have been due to the scrolling between image and question or short term memory. Another possibility is peoples association of the action of increasing with a certain directional bias. People would associate making darker with increasing and thus may ignore numbering layout/convention all together, those who answered clockwise if things were consistent would have moved the dial to 0.
Question 2 and 3 in section one was also inconsistent in response patterns. People almost universally expected to have to turn the dial the opposite direction than what they answered in the previous question despite the fact that the image was just flipped 180 degrees meaning the action would have been the same if things were mechanically consistent. This pair unlike question 1 and 6, people may have had the action still in their memory and so when they read move the dial up, were primed to consider moving the knob the other direction than what they had previously answered for a down action. People who answered clockwise were more likely to change their answer as only one person was consistent between both for clockwise while with counter-clockwise five people stuck with the same direction of movement, though I don’t think this pattern would hold up over a larger sample size.
From the data gathered it seems the most consistent place to put a knob which would control a dial is to the left of it as it was the only question with universal responses. Another design pattern that would appear best is having the dial be horizontal as people were far more consistent in their responses when the dial was horizontal than on the vertical examples (50-60% consensus vs 80-90% consensus.) Another thing to consider would be that understanding number conventions on knobs was very inconsistent. People tended to ignore conventions between the examples and had a preconceived idea for which direction would “increase.” From these examples it was also clear that it was not consistent whether numbers increasing or decreasing meant increasing the effect. The photocopy example was nearly 50% consensus and so was the obscured number dial so it seems it cannot be assumed that people associate larger or smaller abstracted numbers with an increase in effect (in this case the photocopier being darker.)
If I were to follow up on this study I would do more work with multiple knobs though on a horizontal pattern as it was the most consistent amongst participants (in a radio example a knob for frequency and volume.) As for knobs that controlled functions like the photocopy example, I would try different numbering conventions, like more specific numbers (percentages, time, etc…) or graphics (a gradient or sloping pattern) and other functionalities that people might use them for (a stove top/oven.) People certainly had no problems understanding how such a device might work though they were very inconsistent with their direction of movement. I think further investigations may help determine things like learned use and comprehension of patterns and consistency between designs.