The secondary task's own presence changes how the primary task is performed — the measurement interferes with what it measures

What it is

The secondary-task technique carries an inherent methodological problem: the mere presence of a secondary task changes how participants perform the primary task. This phenomenon — the measurement itself interfering with what it measures — is sometimes called reactivity. Primary-task performance measured with a secondary task added may no longer be the same as how the participant would naturally perform the primary task alone.

Why it happens

Participants know they've been asked to do two things at once, and even when explicitly told to prioritize the primary task, being "monitored and split" still pushes people to change their usual operating pace, checking frequency, or attention allocation habits — speeding up to free up time for the secondary task, or deliberately suppressing responses to the secondary task to protect the primary one. Neither of these adjustments would occur if the participant were doing the primary task alone; they're induced by the dual-task situation itself. This means the "primary-task load" measured through this method is, to some degree, mixed together with the extra pressure that comes from simply being asked to handle something else at the same time.

Studying it

A common way to check for reactivity is to run two conditions — the primary task alone (a baseline with no secondary task) and the primary task with the secondary task added — and compare whether primary-task performance or operating style (keystroke rhythm, checking frequency) differs systematically between them. If primary-task performance differs noticeably between the two conditions, the introduction of the secondary task has already changed how the primary task is executed, and any later interpretation of secondary-task performance needs to account for this contamination. Methodologically, it's a mistake to assume that unchanged primary-task performance numbers alone prove there's no interference — a change in operating style (slowing down, altering scan patterns) may not show up as a drop in the performance number, yet it equally means the secondary-task technique isn't capturing the participant's naturally occurring load.

Where it stops holding

Reactivity is more pronounced in situations where the primary task's pace is flexible and participants have real room to self-adjust — self-paced reading or browsing, for instance. In situations where the primary task's pace is externally locked and participants have no room to adjust their operating style — a fixed-cadence tracking task, say — the interference from reactivity is comparatively limited, and the secondary-task method's readings come closer to the true state.

Related

• Same group: A9.08.1 Unchanged primary-task performance does not mean load hasn't increased — resources may simply have slack left · A9.08.2 The secondary-task method infers the primary task's remaining resources from the performance drop on an added task · A9.08.4 The secondary-task method is sensitive when the two tasks compete for the same resource, and underestimates load when they don't
• Adjacent: A9.02 Measuring Load
• Search terms: reactivity · measurement intrusiveness · dual-task interference