Body reading
EEG and event-related potentials
EEG / ERP
Measures: Cognitive effort over time, and, in the ERP cut, the reaction to a single event
If someone says the new flow “feels lighter”, EEG is the kind of evidence that turns that into a number. And there is a second, finer use that hardly ever comes up in design conversations: cut around an event, it shows the millisecond at which a label failed to match what the person expected.
EEG picks up the brain’s electrical oscillations through sensors on the scalp, and two of those oscillations change predictably as the task tightens: theta in the frontal midline region rises and alpha in the parietal region falls, in the pattern KlimeschReviewed how alpha and theta rhythms track memory performance, and is the updated source of claims previously misattributed.One reference in this work:1999EEG alpha and theta oscillations reflect cognitive and memory performance: a review and an…See in the bibliography → described in 1999. That pair is the most widely used index of mental load, and it is what makes it possible to say, with a number, that one screen demanded more than another. GannouniUsed EEG and machine learning to detect emotion in usability testing, and is the updated source of a finding previously misattributed.One reference in this work:2023Software usability testing using EEG-based emotion detection and deep learningSee in the bibliography → and colleagues took the signal into software usability testing in 2023, classifying the emotional state of the people using it from the electrical reading, with high accuracy within their own experimental design, which is where that kind of number holds; and in a 2020 study with nineteen people exploring the rooms of an automated house, AngiolettiMeasured physiological response to smart-home systems, and is the correct source of a finding the original work attributed to another author.One reference in this work:2020Neurophysiological correlates of user experience in smart home systems (SHSs)See in the bibliography →, Cassioli and Balconi measured EEG and autonomic signals together, and neural activity changed consistently between being at rest and interacting with the environment, and the most multisensory room was the one that produced the greatest emotional response and increase in heart rate.
The second use swaps accumulated load for a surgical cut: instead of looking at the whole session, you isolate the window of signal around an event — the instant the screen shows the label, the icon, the price — and average many repetitions. The result is the event-related potential. The most useful component for anyone who writes interface copy is the N400, which appears about four tenths of a second after a word that does not fit the context, and its size tracks the size of the mismatch (KutasDiscovered the N400, the electrical peak that appears when the brain meets something that doesn’t fit what it expected.One reference in this work:1980Reading senseless sentences: brain potentials reflect semantic incongruitySee in the bibliography → and Hillyard, 1980). It is the electrical signature of a label that betrays expectation, and it is the kind of discussion that, in product, usually gets settled by guesswork.
SeesTemporal resolution in milliseconds; it detects the exact instant of friction.
Does not seeLow spatial resolution, sensitive to movement artefacts, and the signal, on its own, does not say which emotion is involved.