I think you're most directly talking about Cognitive Load, the term in cognitive psychology, but also in Human Computer Interaction. We have a finite amount of working memory and the more complex an interaction or interface is the more difficult it is to process it.
Here's a good scholarly overview on Cognitive Load and Complexity in HCI. Complexity is a good general term to use to describe the effect you're talking about.
The design of media must be done in a way, which does not overstrain our
cognitive abilities. An example for supporting the human mind would be using world
semantics, like in Figure 4 , where a virtual trashcan is used when deleting files or the
color red, which in real life also makes us attentive, for an error.
A poorly designed interface that does not take advantage of people's internal model of how the world works will result in higher cognitive load, as they may be left wondering exactly what a term or action means. The virtual trashcan example makes it clear how deleting a file means, whereas using no visual metaphor will result in a more abstract understanding of the issue, resulting in a higher cognitive load.
Something actually moving slower-as you said Firefox is "slow"-isn't quite the same as a cluttered interface in terms of effect on Working Memory, but delays between an action and the result can be uncomfortable and unnatural, causing the user to think "Did X happen? Is it working? Should I click again?" this can distract the mind from the task at hand and certainly consumes working memory, and generally makes an interaction less pleasant.
It's important to note that cognitive load is more complex than Miller's Plus or Minus Two. As is the case with much psychological research, 7+-2 (that is, the idea that people generally hold seven things, give or take two, in short-term memory at once) caught on because of how tantalizing and easy to apply the concept was, not because of how accurate or useful it was. Remember that there is much more to Cognitive Load:
In IR, the concept of cognitive load rarely extends beyond the ideas presented by Miller (1956). In Miller’s famous paper "The magical number seven plus or minus two", a human’s capacity for processing information was explored. It was concluded that short-term memory (working memory) has a limited retention. The study by Hu, Ma, and Chau is typical of much research in IR that advocates attempts to minimise cognitive load during interface design by recognising the limitations of working memory.
The important thing to remember about cognitive load is it should be minimal but appropriate to the task at hand. 7+-2 is not a magic number, but a simple interface can work like magic.