UX Laws: How People Choose
Transcript
Made automatically from the video's sound, so it may contain mistakes. Click a line to play from there.
Welcome back, vibe coders, designers, and product managers. Let's jump straight into today's explainer on how people actually evaluate options, forage through complex navigation, and ultimately commit to choices under uncertainty.
Now, just a super quick recap of our last explainer. Why interfaces feel easy or difficult. We successfully separated those predictive mathematical models from the psychological effects across Norman's gulfs of execution and evaluation.
So, with that foundation laid, we're shifting our focus to the psychological realities of evaluation and choice. Literally, how do people forage through navigation hierarchies? And what makes them decide what to actually click?
Here's our high cent roadmap for today. We'll cover the science of choosing, decoding Hicks Law, busting the choice overload myth, demolishing the three-click rule, comparing broken versus repaired pathways, and finishing up with a dual refactor intake exercise.
Right, let's get into section one, the science of choosing, focusing on mental effort and working memory. So, cognitive load is absolutely the foundation of user decision making. Back in 1988, John Sweller proposed the cognitive load theory, which basically divides our mental resources into two main buckets.
First, you've got your irreducible intrinsic task complexity. That's just the natural, unavoidable difficulty of whatever the user's trying to do. But then you have avoidable extraneous UI clutter. This is the visual noise, the confusing layouts and those unnecessary steps that we, yes, us designers, accidentally add in.
You see, when this extraneous UI noise exceeds our working memory, users simply give up and abandon the task entirely. The takeaway: we have to ruthlessly eliminate the extraneous stuff so our users can actually focus on the intrinsic.
Decoding Hicks Law. We're looking at mathematical models versus folk myths. Now, this principle is so often severely misunderstood by well-meaning vibe coders. Grounded in MIT human factors experiments by Hick in 1952 and Hyman in 1953, this is a strict mathematical formula.
It calculates subsecond motor latency for equiprobable choices, factoring in Shannon entropy weighting. Think of it like a user pressing a physical button, the split second a light flashes.
It is a highly practiced sub-second stimulus response model. It absolutely does not predict the total collapse of complex product evaluation. And this brilliantly illustrates the point. Hicks' law is a strict tier one mathematical model of practiced motor latency.
It is definitely not a blanket justification to hide absolutely everything inside cryptic hamburger menus. When you improperly scale a sub-second reaction model into a tier two cognitive myth to govern complex interfaces, you're just damaging the user experience.
You're basically forcing users to play a frustrating game of hide-and-seek with your navigation. The choice overload myth and what the data actually says. This is where empirical data utterly destroys the popular narrative around choice paralysis, which you probably heard called the choice overload myth.
Across 50 independent studies involving exactly 5,036 participants in a massive 2010 meta-analysis by Scheiberheni, Griffinader, and Todd, researchers tested this hypothesis to see if more options truly paralyze users. And the result?
A mean effect size of a mere 0.02. Statistically speaking, that is virtually zero. So that sweeping generalization that more choices always equals less motivation is quite frankly a complete myth.
What the science actually proves is that large assortments only paralyze users under some highly specific boundary conditions. We're talking about situations where options are uncategorized, the domain is highly unfamiliar to the user, the decision is high stakes, and there's a distinct lack of comparable trade-offs.
What does this mean for us? It means smart categorization completely neutralizes the threat of overchoice. If you organize your complex information well, your users simply won't be overloaded.
Demolishing three-click rules. Let's talk about information sent and foraging. Listen, arbitrary click limits are actively harming your user experience metrics. The whole three-click rule assumes users will just leave in frustration if they can't find their goal instantly.
But, as the Nielsen Norman group unequivocally states, the three-click rule is an arbitrary rule of thumb that is not backed by data. Exhaustive usability testing actually proves that users don't just quit after three clicks.
They confidently keep clicking, provided they maintain the scent of their goal. So we need to replace click counting with information sent. This is an essential concept from Paroleon Card's information foraging theory. Think about it.
Humans forage for information online in much the same way animals forage for food in the wild. Clear, specific labels create a high information scent. If that scent is strong, users will happily click ten times.
But if the scent is weak, they'll abandon the task on click number one. Your job is to design the scent, not to sit there counting the clicks. Broken versus repaired pathways. Let's look at multimodal accessibility design.
Let's expose this common AI UI trap of broken pathways. Failure mode A is just dumping raw database rows into an ungrouped 48 course internal jargon wall. It's a nightmare.
Failure mode B is blindly obeying that Hicks law minimalism by forcing users through an overpruned mystery wizard, you know, asking them to pick paths like catalyst, pioneer or weaver, and usually sneaking in a deceptive pre-checked 49 certification boost add on.
Both of these approaches cause massive working memory exhaustion. On the flip side, a repaired pathway, like a scannable, categorized mega menu with progressive disclosure, actually respects the user's time and intent. And a truly repaired pathway safely accommodates all strict accessibility modalities without resorting to unlabeled card soup.
On mobile, we use collapsible category accordions. For keyboard and switch users, we wrap the navigation in a native nav ARIA label tag, use native ARIA expanded trigger buttons, a logical tab order, and crucially an instant escape key bypass, so they never have to involuntarily tab through 48 different links.
For Zoom, we guarantee a 400% vertical reflow down to 320 CSS pixels with no truncated fee disclosures or horizontal scroll bars. And for screen readers, we implement semantic H2 and H3 category landmarks alongside explicit table header scope cells.
Finally, section 6, the dual refactor intake exercise. Because context is absolutely everything. To really drive this home, let's mentally refactor a bloated 19-field scholarship intake screen.
You see, a single interface simply cannot serve two vastly different cognitive needs. For an anxious first-time learner on a mobile phone, or adhering to W3C Koga standards, a GovUK style one thing per page design absolutely wins every time, keeping that extraneous cognitive load to an absolute minimum.
However, that exact same pattern would infuriate an enterprise worker. Dense single screen multi-attribute tables are absolutely required for practice daily staff who are reviewing 100 applications a day. They literally need high cent triage on a single screen to do their jobs efficiently.
These are the crucial edge cases you have to map out in your daily work. You must always match your design pattern to the user's expertise and their environment. That's how you safely manage intrinsic complexity while brutally eliminating extraneous UI clutter.
And as a final, non negotiable ethical guardrail mandated by user trust, you must employ transparent, revocable smart defaults that genuinely help the user, rather than relying on deceptive, opt-out, dark patterns designed to trick them. So I'll leave you with this final provocation to chew on.
If Higgs Law doesn't demand absolute minimalism, and choice overload is easily mitigated through smart categorization, how will you redesign your interfaces to maximise information sent instead of just blindly minimizing clicks? Thank you so much for joining me for this explainer.
This perfectly sets the stage for our next one, Series 11 episode 2, where we'll uncover how people act targets, movement, and control. See you then.