Plain-language explanation.
Attention is the ability to focus on what matters and ignore distractions. We cannot process everything our senses detect, so attention selects what to prioritise. Different types of attention exist: selective (picking one thing to focus on), divided (attending to multiple things), and sustained (maintaining focus over time).
Core concepts and standard treatment.
Selective attention theories (Broadbent's filter theory, Treisman's attenuation model, late selection models) explain how the brain selects relevant from irrelevant information. Top-down (goal-driven, frontoparietal control network) and bottom-up (stimulus-driven, saliency) attention systems interact. The attentional blink reveals temporal limitations in conscious processing. Spatial attention (Posner cueing paradigm) is coordinated by frontal eye fields and intraparietal sulcus.
Deeper theory, debates and edge cases.
Neural substrates: the dorsal attention network (frontal eye field + intraparietal sulcus) implements goal-directed top-down attention; the ventral attention network (TPJ + right prefrontal cortex) implements stimulus-driven reorienting; default mode network deactivates during externally directed attention, and its failure to deactivate is associated with mind-wandering. Neuromodulation of attention: acetylcholine enhances signal-to-noise ratio; noradrenaline regulates arousal and phasic responses to salient stimuli.
How it is applied in practice.
Applied attention research: ADHD involves dysregulation of prefrontal-striatal-cerebellar circuits — first-line treatments (methylphenidate, amphetamine) enhance dopamine and noradrenaline signalling. Mindfulness training strengthens sustained attention and reduces mind-wandering. Technology and attention: smartphone notifications create continuous partial attention; social media design deliberately exploits bottom-up attentional capture through variable reward schedules. Educational design principles informed by cognitive load theory (Sweller) minimise extraneous cognitive load while maximising germane processing.