Interesting. I’ve seen Gestalt referred to in UX but know it better from its therapeutic history. I get where you’re coming from, and how it may possibly be applied to UX as an over-arching set of principles. But I’m not convinced that they’re either necessary nor applicable. And for these reasons. Would be curious what you think:
- I don’t find the proximity example convincing. First, the example uses both proximity and likeness, so it’s hard to visually separate the grouping that we do (cognitively) because the individual elements are visually identical. So then there’s a confusion in the example of a distinction between pattern and proximity. Since the example includes repeating elements, I’d need to see different examples of proximity to see how Gestalt makes its point.
- The closure example, I think, is unconvincing. I personally dont think Gestalt means that we seek visual closure, but rather some other kind of cognitive closure. Which I dont think then would apply to navigation of a partially disclosed and partially concealed navigation schema. I dont need to scroll all the way left and all the way right to obtain a sense of closure. And I dont find my own navigation or behavior on apps and sites to be driven by any need of functional closure — not sure bout others. But whereas I can see closure being an interesting feature of interpersonal exchanges, for example, or relationships, dramatic tellings (endings), etc. I dont know that it applies to navigation.
- I would tend to view layout and design options through a combination of approaches, but among them I’d certainly use content strategy, IA, interaction design, design and UX patterns, and other aspects of what all together comprise a familiar experience for users of a combined information space, interaction domain, practice, and of course content. Content in itself is a huge one — we do read UI after all, drawing both on our familiarity with the visual composition and content architectures and styles of the publishing world, as well as idioms of application and interface types, as well as of interface elements and features (we know IOS, web, etc now well enough to read them without much cognitive effort), as well as of course linguistic meaning itself (text). Here seems to me Gestalt has no application whatsoever, since it emphasizes basic structures or forms over learned and habitual experience with content and interaction and technology domains that we become used to over time (and often very quickly).
- Running longer in this comment than I expected, so that obviously means this raised some interesting points. But finally, I think there’s something to the notion of conceptual forms that maybe could be teased out in relation to what makes “good” UX. And applied maybe through heuristic evaluation. That would be worth exploring. I’ve always been a bit underwhelmed by the explanation of the criteria we apply in heuristic evals, and think something like this might be closer to the truth than some of the explanations we’ve all seen for what makes a good experience (e.g. search > results = information. But maybe closure works here better?)