We're working on a filter interface. Most options can be represented by two options, one using "hide" as the primary verb, and the inverse formulation using "show" as the primary verb.

Discussion on the team and with users has been largely based on personal preference. I'm wondering if there are any UX studies or behavioral economics principles that demonstrate quantifiable effects from using one versus the other. "Show", to me, sounds more active and positive, while "hide" sort of implies that the tools is failing me by showing me irrelevant things...but I don't know if this is my own opinion or representative of an actual UX principle at play. There's also been an argument that we're showing you a set of items by default, and the "hide" terminology more accurately reflects that we're taking out some of what was previously included in that list.

I'm aware of the framing effect, but I'm having trouble figuring out whether using the verb "hide" will frame these options as a loss and therefore contribute to a sub-par experience.

Here's what the filter looks like. (We've decided that to make parsing the options easier, we're probably going to start them all off with the same verb.)

homepage filter screenshot

Taking the last option as an example, should I expect any difference in behavior if I use "Hide questions with score < X" rather than using "Show questions with score > 0", even though they mean the same thing?

  • 3
    Personal anecdote, vs. hard numbers: I always prefer to frame checkbox options positively, instead of negatively, to avoid confirming a negative. I'd be hard pressed to successfully argue that "hide" is truly negative, but it just sounds more negative to me. Provided the action's verbage allowed for it, I would always lean towards "show" as a first reaction. Commented Nov 19, 2014 at 20:57
  • My answer isn't UX based, so I'm listing here as a comment. From a programming standpoint, I favor show buttons in a situation like this, because it empowers the user to add to the report, rather than reduce it. This allows the initial, pre-filtered version of the report to contain fewer-lines/less-data which usually equates to a shorter rendering time. A report which shows up instantly with some of the desired data, which then invites the user to add more detail as need,... now that is a positive user experience to me. Commented Nov 19, 2014 at 21:20
  • Imagine the words “Hide” and “Show” were not there. The rest being unchanged, what would a checked box mean? The last two entries in your example should probably not have checkboxes at all because the conditional takes care of that.
    – Crissov
    Commented Nov 19, 2014 at 23:46

1 Answer 1


Looking around there are not any studies I can see as such. Something I would rely on from experience would be consistency. The majority of interfaces use this terminology to describe the action and users understand this.

Alternatively you could follow this approach:


You do not explicitly say the word hide, but the checkboxes imply it.

It would largely come down to preference in this case, studies do show that it is preferable to use positive language. In this case however, the user may see the action of hiding 'unimportant/irrelevant' information to be positive and may not even register any negative connotations from this.

Also using the sign '>' is requiring a user to think more than they need to. People who are not familiar with this notation may be confused and it would be advisable to use natural language instead.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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