Short answer

Replace vague navigation terms with high-scent labels that explicitly match the vocabulary of your user's primary goals.

Field: Commercial Production
Source: ACM Transactions on Computer-Human Interaction (2015)
Method: Automated Computational Cognitive Modeling (Simulation)
Evidence: Strong effect

Automated cognitive modeling reveals that high semantic similarity between a user's goal and link labels minimizes the probability of 'lostness' by streamlining the scent of information. This commercial production research insight is drawn from a 2015 study published in ACM Transactions on Computer-Human Interaction. Using Automated computational cognitive modeling (simulation), researchers explored how this design variable affects real-world outcomes. The key design takeaway: Replace vague navigation terms with high-scent labels that explicitly match the vocabulary of your user's primary goals.

Study

Commercial ProductionHigh ImpactStrong effect

Semantic label consistency in Information Architecture reduces user navigation failure and cognitive load

Automated cognitive modeling reveals that high semantic similarity between a user's goal and link labels minimizes the probability of 'lostness' by streamlining the scent of information.

ACM Transactions on Computer-Human Interaction · 2015

Key Findings

01Improved labeling systems significantly reduced the number of navigation steps required to find information
02Enhanced semantic scent accounts for a 20-30% reduction in 'lostness' metrics within deep site hierarchies
03Database-oriented modeling can identify 'dead-ends' in IA before a site goes live

Application

Design takeaway

Replace vague navigation terms with high-scent labels that explicitly match the vocabulary of your user's primary goals.

How to apply

Analyze your existing site menu using a 'closed card sort' to see if users associate your labels with the correct content. If more than 20% of users fail to match a label to its destination, rewrite the label to be more descriptive.

Project actions

01Test your website's navigation tree (Tree Testing) before you ever start the visual UI design
02Keep labels consistent across different levels of the website
03Avoid jargon that your target audience wouldn't use daily

Method & Evidence

AimCan a computational cognitive model accurately predict and improve navigation performance by optimizing Information Architecture (IA) labeling systems?

MethodAutomated Computational Cognitive Modeling (Simulation)

ProcedureResearchers mapped the project of two large websites into a database and simulated user navigation using cognitive agents. These agents attempted to find specific information goals by evaluating the 'scent' of links. The labeling systems were then iteratively optimized for semantic clarity, and the simulations were re-run to compare success rates.

ContextLarge-scale corporate and educational website information architectures

Variables

IVSemantic clarity and consistency of IA labeling systems

DVNavigation success rate (measured by simulated user goals achieved) and cognitive load (implicitly represented by agent decision-making processes and efficiency)

CVWebsite information architecture structure, complexity of information goals, cognitive agent parameters (e.g., information scent evaluation algorithm)

Strengths & Limitations

Strengths

+Utilizes a computational approach allowing for large-scale simulations not feasible with human participants.
+Investigates a clearly defined aspect of IA (labeling) with systematic optimization.
+Provides a quantifiable measure of navigation performance and cognitive load through simulation metrics.

Limitations

As a student, you likely won't have access to complex 'computational cognitive models,' so you must use human participants to simulate these results through manual testing.

Reliability & validity

The study's computational nature enhances reliability through consistent application of algorithms. Validity is strengthened by simulating a known user behavior ('information scent') and measuring quantifiable outcomes. However, it may lack ecological validity due to the exclusion of visual cues and real user variability, potentially oversimplifying the cognitive processes involved.

Think critically

Does an 'optimized' label always lead to a better brand experience, or can some descriptive labels make a website feel clinical and boring?

Design Principles

"Information Scent Optimization"

Users rely on 'information scent' to make split-second navigation decisions; when labeling is ambiguous or inconsistent, cognitive load spikes, leading to site abandonment. By mathematically aligning label vocabulary with user mental models, designers can ensure predictable navigation even in high-complexity environments.

What This Means for Your Design

If the names of the buttons on a website don't clearly match what a person is looking for, they will get lost. Using better words makes the website much easier to use, which can be proven with computer simulations.

How to use in your project

1.Use this to justify why you changed a navigation label from 'Resources' to 'Student Study Guides' during your iterative design process
2.Reference the concept of 'Information Scent' when discussing your Information Architecture

Add to My Project

Quick Cite

Paragraph starter

According to Van Schaik et al. (2015), optimizing the labeling system within an information architecture can significantly reduce navigation errors by strengthening the 'scent' of information for the user.

Source

ACM Transactions on Computer-Human Interaction

Automated Computational Cognitive-Modeling

journal · 2015

View source

Questions About This Research

What does the research say about semantic label consistency in information architecture reduces user navigation failure and cognitive load?: Replace vague navigation terms with high-scent labels that explicitly match the vocabulary of your user's primary goals. Evidence: ACM Transactions on Computer-Human Interaction (2015).
Why does "Semantic label consistency in Information Architecture reduces user navigation failure and cognitive load" matter for design?: Users rely on 'information scent' to make split-second navigation decisions; when labeling is ambiguous or inconsistent, cognitive load spikes, leading to site abandonment. By mathematically aligning label vocabulary with user mental models, designers can ensure predictable navigation even in high-complexity environments.
How can designers apply this research?: Replace vague navigation terms with high-scent labels that explicitly match the vocabulary of your user's primary goals.
What were the main findings?: Improved labeling systems significantly reduced the number of navigation steps required to find information. Enhanced semantic scent accounts for a 20-30% reduction in 'lostness' metrics within deep site hierarchies. Database-oriented modeling can identify 'dead-ends' in IA before a site goes live
What research method was used?: Automated Computational Cognitive Modeling (Simulation).
How strong is the evidence?: Evidence strength is rated Strong effect, based on a 2015 journal from ACM Transactions on Computer-Human Interaction.
What should I do differently in my next project?: Analyze your existing site menu using a 'closed card sort' to see if users associate your labels with the correct content. If more than 20% of users fail to match a label to its destination, rewrite the label to be more descriptive.
What are the limitations?: The model primarily focuses on text-based labels and may not fully account for the impact of visual hierarchy, icons, or emotional design elements on navigation.