5
$\begingroup$

Say we A-B tested user reactions to completing some task (e.g. transferring files, restarting their device): would people be quicker to get impatient/ annoyed if they got random percentages as indicators of progress, or if they had a steady, accurate progress bar? What if it was steady but NOT accurate? When would people experience the most intense emotional drops or rises? Would the slot machine effect play a part, i.e. would the chance of getting a "big" leap (21% towards update installation!) keep users waiting patiently during the slower progress intervals?

This question is based off a wandering thought while my laptop was restarting recently, as well as some basic psych understanding. I can certainly see it being possible that people would, on average, be more willing to wait for a 2/3/8 min update with erratic progress than 1/2/5 mins with steady progress. I'd be interested if anyone knows of such/similar studies and, if so, what conclusions they reached.

Edit: internet legend Tom Scott came through and discussed this topic recently, https://www.youtube.com/watch?v=iZnLZFRylbs!

$\endgroup$
4
$\begingroup$

A couple of studies (Harrison et al, 2007; Amer & Johnson, 2016) tested pauses/stalls in progress, and concluded (respectively): "... participants perceived progress bars with pauses as taking longer to complete ..." and "... participants preferred a progress bar that moves consistently to a stalling progress bar, and judged the process duration to be shorter ..." suggesting that continuous/steady progress bars are preferred by users over intermittent/random indicators.

There is much more research on continuous progress indicators (eg, Villar, Callegaro, & Yang, 2013; Liu & Wronski, 2018) that suggests that users generally prefer fast-to-slow / decelerating progress - that is, inaccurate progress bars that overestimate progress early on. Slot machines similarly use early success to increase player perseverance, a kind of anchoring bias. However, most of this research is on progress indicators for active tasks such as filling out surveys. Some experiments with passive waiting (like file transfers) align with their results (eg, Gronier & Baudet, 2019; Kuroki & Ishihara, 2015), but other studies show a preference for accelerating progress instead (eg, Harrison et al, 2007), so I would wait for a meta-analysis specific to passive tasks before making a firm conclusion.

There are also various exploratory studies testing alternative progress bar displays, such as different colours and shapes, designs and animations, adding audio and video, suggesting that user perceptions can be altered in a variety of novel ways. As well, progress indicators on passive tasks may be affected by the labour illusion, in which user experience is influenced by the perceived amount of work done.

$\endgroup$
2
  • $\begingroup$ is this a kind of lying? What are some ethic aspect of this? $\endgroup$
    – Ooker
    Oct 21 '20 at 14:20
  • 2
    $\begingroup$ @Ooker You may be interested in this article - there is much worse going on with progress bars, and it's all done because users prefer it that way! $\endgroup$
    – Arnon Weinberg
    Oct 21 '20 at 17:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.