Attentional Control

What it is

The cleanest demonstration of what the construct is comes from memory competition. On a working memory task, competitors hold about seven items where a typical student holds around two, and the difference lies in attention rather than in memory itself. They are better at keeping interference out, not at storing more.

That case is worth stating carefully, because a popular account of it reverses the arrow. Reading the competitors' advantage as memory training producing attentional control is the inverse of what was observed.

The construct's most cited empirical test is Eyal Ophir, Clifford Nass and Anthony Wagner's 2009 paper, which screened 262 participants and compared heavy and light media multitaskers on filtering, working memory and task switching. The heavy group did worse at filtering irrelevant information. The first author is Ophir, then a graduate student, and the vault's standing rule is to credit him rather than the better-known second author.

In effect

The 2009 finding travelled a very long way. It reached most readers through a radio interview in which heavy multitaskers were described as pretty much mental wrecks, in language of lasting effects and rewiring, and popular accounts have extended it to the reader's own brain: glance at your phone at every moment of boredom and your brain has likely been rewired.

The paper underneath cannot support that. It is correlational and cross-sectional. It compares two groups of people at one moment, with no longitudinal data, no intervention, no imaging and nothing about causal direction. It cannot tell you whether multitasking degraded anyone's filtering or whether people who already filter poorly are the ones drawn to multitasking. Lasting, rewired and negative effect are all causal and temporal claims a snapshot cannot make.

The construct itself is not in trouble. If you want to argue that switching between tasks is costly, the better evidence is switch cost and attention residue, which measure the cost directly rather than inferring it from a group difference.

What it does not say

It does not say that heavy media multitasking impairs attention. Wiradhany and Nieuwenstein ran two close replications using the original experimental program and the original questionnaire, and both failed to reproduce the association. Their meta-analysis of 39 measures across 12 studies found only a small association that was not robust, and a ten-years-on meta-analysis reaches the same verdict.

It does not say the original researchers overclaimed. Uncapher and Wagner, by the original study's own senior author, is a careful review conceding that the literature is mixed. That is the citation to reach for, because it comes from the side that would most like the effect to be real.

It does not say the construct is dead. The headline is. Describe the 2009 paper as historically important and empirically shaky, and make the case against multitasking from the measures that test it directly.

A note on attribution. The radio phrasing belongs to a scholar who died in 2013 and cannot answer for it, which is a reason to criticise the claim and not the man. The archived segments that could be reached are dated 2009 and 2013, and a 2010 interview could not be confirmed, so check the endnote before printing a date.

---