Fil Salustri's Design Site

Site Tools


Interaction Error Chart

An Interaction Error Chart is a simple representation of a single undesired event occurring during a user's interaction with a product.

This page is no longer used. You probably are looking for the interaction error page.

What is an IEC?

An interaction error happens when a product cannot accommodate a user's need. The error may arise from a lack of functionality, a lack of affordance for the particular user, a lack of general usability, or a lack of appropriate feedback to complete the interaction. Interaction errors happen only at the interfaces between interacting systems; for our purposes, we can further restrict them to only those errors when a human and a “machine” interact (per the HMIL).

We represent each interaction error with a simple chart. These Interaction Error Charts (IECs) are only meaningful with respect to appropriately defined personas, a SUC, and a specific step in a usage scenario.

IECs only describe the error, not how the error may be addressed by (re)design work1)([GM98], [MGW98]).

How do we build an IEC?

An IEC is a simple two-column chart that models a single undesired event occurring during an interaction between users and a product in a given SUC. This implies the following process:

  • For each SUC:
    • For each usage scenario consistent with the SUC:
      • For each step or error of the usage scenario:
        • For each interaction error you can imagine could reasonably happen to the personas in the SUC:
          • Construct a corresponding IEC.

Besides specifying the US step, the SUC, and the specific interaction error, three other aspects must be described.

Relevant human factors
Identify by name/type the specific human factors that could come into play for this specific error, for the specific personas, and in this specific circumstance.
Relevant design features
Identify the relevant features, aspects, functions, or behaviours of the design that pertain specifically to this interaction error. If a reference design is the subject of the analysis, specific features of that reference design can be mentioned explicitly.
Persona responses
Describe briefly how you expect each participating persona to react to the error, and the actions those personas might take to address the interaction error.

A sample IEC is shown in Table 1.

Table 1: A sample IEC for the elevator case study.
US STEP: 2.2a.2/1 User requests to go up.
SUC: 1: Aisha & Lars go to work.
ERROR: Elevator fails to provide feedback confirming request to go up.
RELEVANT HUMAN FACTORS: Perceptual: the elevator’s noise is additional (unintended) feedback about its direction of travel; the floor indicator can tell them if the elevator is coming.
Cognitive: Can users distinguish cognitively between the sounds the elevator makes? How did they come to know the distinction? Can they deduce the direction of travel from the floor indicator?
Motor/Physical: Aisha feels discomfort in her neck, shoulders, arms, and hands from carrying the heavy bag.
Psychosocial: Lars is worried about being late for his first meeting.
RELEVANT FEATURES: Location of elevator interface, brightness & size of call button light, location of floor indicator; ambient noise (that may mask elevator’s noise).
RESPONSE: Lars takes the stairs. Aisha hears the elevator moving and does not want to carry the exams up four flights of stairs, so remains in hope that it’s only the feedback light that is broken and that transport will arrive. Aisha does not know Lars well enough to be comfortable asking him to help carry the exams up the stairs.

Number IECs sequentially from 1. This lets you easily refer to them in subsequent text by ID (e.g., “IEC #5”).

IECs are failures of interactions, not failures of product function. However, interaction failures often occur as a result of a failure of product function.

  • This is (probably) the case in Table 1.
  • The interaction error is a feedback failure.
  • Such a failure could result from a functional failure in the elevator.
    • Example: an indicator light burns out.
  • Such a failure could also result from a non-functional product failure.
    • Example: the elevator fails to provide feedback in a form that the user can receive.


Expected deliverables include:

  • A set of IECs covering a representative set of interaction failures for all SUCs, USs, and personas.
  • Up to 250 words of explanatory text for each IEC, explaining anything noteworthy or peculiar about the IEC.


This could lead to an enormous number of IECs. For instance, if you have five USs, each with three SUCs, and that each US has ten steps or errors, you would likely have a minimum of 150 IECs.

While in “real life” design situations you would very likely be expected to described them all, this is not possible under the constraints of a typical academic course.

You are therefore expected to find a representative subset of IECs that capture the breadth of the range of IECs and also represents the typical and also most significant interaction errors.


GM98. a J.S. Gero and T. !McNeill. 1998. An approach to the analysis of design protocols. Design Studies, 19:21-61.
MGW98. a T. McNeill, J.S. Gero and J. Warren. 1998. Understanding Conceptual Electronic Design Using Protocol Analysis. Research in Engineering Design, 10(3):129-140.
The theory behind this is that the human brain works best when it can focus on either analytic work or design work, but not both in rapid alternation.
design/interaction_error_chart.txt · Last modified: 2020.03.12 13:30 (external edit)