Attention Residue

Attention residue was documented by organisational psychologist Sophie Leroy in a 2009 paper that set out to understand why task-switching is so cognitively costly. The established finding was that switching between tasks takes time — there is a measurable switching cost in terms of reaction time and error rate. Leroy's contribution was to identify a different, and in some ways more serious, cost: when you switch away from a task before it is complete, part of your cognitive resources remain on it. The previous task leaves a residue in working memory that competes with processing the current one.

Leroy demonstrated this through experiments in which participants switched between tasks under different conditions. Those who switched while a task was still unresolved showed significantly degraded performance on the subsequent task compared to those who had completed or clearly paused the previous one. The residue was not a metaphor — it was measurable in performance outcomes, and it persisted beyond what simple fatigue or switching costs could explain. The mind, it turned out, does not switch cleanly. It lingers.

The connection to smartphone use follows directly. A phone check during cognitively demanding work is a task switch of a particular character: you are moving from a complex, effortful cognitive task to a social and informational stream specifically designed to capture and hold attention. The social content of the phone — messages, notifications, updates — is processed emotionally and socially, not just informationally. When you return to the original task, you carry cognitive and emotional residue from that social content. A message containing mild interpersonal friction, a news headline that provokes anxiety, a post that triggers social comparison — these do not stay in the phone. They follow you back.

The practical implications are different from what most productivity advice implies. The conventional framing treats phone interruptions as time costs: the check takes two minutes, so you lose two minutes. Attention residue reveals the cost as dimensional rather than temporal. The two-minute check may reduce the quality of the following thirty minutes of thinking, because the residue occupies working memory that would otherwise be available for the task. An hour of genuinely uninterrupted work is not equivalent to two hours of work punctuated by phone checks, even if the total time on task is the same.

Leroy also found that the residue effect is worsened when tasks are left in a state of unresolved activation — when there is something unfinished that the mind is tracking. This suggests a specific relationship between the social stream of the phone and residue severity: social interactions are by nature unresolved. A message invites a reply. A post invites a reaction. A notification implies an obligation. Each check opens social loops that the mind continues to process after the phone is put down, because these loops carry social stakes that the brain treats as requiring resolution.

The intervention that follows is not simply reducing phone use but managing the completeness of transitions. Before beginning cognitively demanding work, clearing open social loops — responding to messages, closing outstanding notifications — reduces the pull of residual activation. Research by Adrian Ward and colleagues found that the mere visibility of a smartphone on a desk depletes available cognitive capacity, even when the phone is ignored entirely. Placing it out of sight removes the cue that triggers task-switching before it can fire. The goal is not perfect discipline but a cleaner cognitive boundary between the social monitoring mode the phone induces and the deep processing mode that serious work requires.