Quantifying attention span across the lifespan

Abstract

Introduction: Studies examining sustained attention abilities typically utilize metrics that quantify performance on vigilance tasks, such as response time and response time variability. However, approaches that assess the duration that an individual can maintain their attention over time are lacking.

Methods: Here we developed an objective attention span metric that quantified the maximum amount of time that a participant continuously maintained an optimal "in the zone" sustained attention state while performing a continuous performance task.

Results: In a population of 262 individuals aged 7-85, we showed that attention span was longer in young adults than in children and older adults. Furthermore, declines in attention span over time during task engagement were related to clinical symptoms of inattention in children.

Discussion: These results suggest that quantifying attention span is a unique and meaningful method of assessing sustained attention across the lifespan and in populations with inattention symptoms.

Keywords: attention span; attentional modeling; continuous performance task (CPT); sustained attention; vigilance decrement.

FIGURE 1

(A) Stimuli and protocol for the CPT. There were a total of 250 trials, with 80% targets and 20% randomly occurring non-targets. (B) Z-scored RTs from an example participant. Each RT was z-scored and plotted over time. RTs that are faster than 1 z-score above the mean are plotted in dark gray and are labeled as “in the zone” trials. RTs slower than 1 z-score above the mean are plotted in light gray and are labeled as “out of the zone” trials. Trials in which there was an error were plotted in red and were labeled as “error trials”. The dashed vertical line represents the break between the first and second CPT blocks. The dotted box highlights the longest period during the CPT when this participant was able to maintain an “in the zone” state (i.e., their A-span).

FIGURE 2

Scatterplots showing that, in young adults, (A) A-span was unrelated to RT and (B) RTV, and that A-span percent change was unrelated to (C) RT percent change and (D) RTV percent change.

FIGURE 3

Age group effects on A-span metrics. (A) Age effects on A-span were driven by children and older adults having shorter A-spans than young adults. (B) Age effects on A-span percent change were driven by children having greater A-span decrements (i.e., a more negative A-span percent change) than young adults. Box and whisker plots represent the bounds of each quartile. Dashed lines represent the group average. White dots represent the group median. Blue significance bars indicate significant interactions revealed from the ANOVAs and black significance bars indicate significant t-test results. *p < 0.05, **p < 0.01.

FIGURE 4

Relationships between A-span measures and inattention in children. (A) There was no significant relationship between the VADRS-IA score and A-span. (B) There was a significant relationship between the VADRS-IA score and the A-span % change. *p < 0.05.