5.1 Theories of Attention

Foundations of Attention

Attention is your brain's system for filtering the massive amount of information hitting your senses at any given moment. Without it, you'd be overwhelmed by every sight, sound, and sensation simultaneously. Researchers have spent decades debating when and how this filtering happens, and those debates produced the major theories you need to know for this unit.

Concept of Attention

At its core, attention is the selective focus on specific stimuli while ignoring others. Your brain has limited cognitive resources, so it can't fully process everything at once. Attention is how it decides what gets priority.

There are three main types of attention:

• Selective attention is focusing on one specific input while tuning out the rest. Think of following a single conversation in a loud, crowded room.
• Divided attention is splitting your focus across multiple tasks at the same time, like driving while having a conversation with a passenger.
• Sustained attention is maintaining focus over a long period, such as staying alert during a two-hour lecture or a security monitoring shift.

All three types serve the same underlying purpose: they prioritize relevant information so you can act on your goals. Attention also supports working memory by keeping task-relevant information active and accessible while you use it.

Early vs. Late Selection Theories

The central debate in attention research comes down to one question: at what point does your brain filter out irrelevant information?

Broadbent's Filter Theory (Early Selection)

Broadbent proposed that attention acts as a filter before meaning is processed. Information gets screened based on physical characteristics like location, pitch, or loudness. Only the information that passes through this filter receives deeper, semantic processing.

• The filter sits early in the processing stream, so unattended information gets very little analysis.
• This model assumes a strict bottleneck: only one channel of information can be fully processed at a time.

Deutsch & Deutsch Model (Late Selection)

The late selection view flips this around. It argues that all incoming stimuli are processed for meaning automatically. Selection only happens afterward, when your brain decides which of those fully processed inputs reaches conscious awareness.

• Unattended information still gets semantic processing under this model.
• The bottleneck occurs later, at the response or awareness stage rather than at perception.

Key evidence for each side:

• The cocktail party effect is often cited for early selection. You can hear your own name spoken across a noisy room, which suggests some physical features of unattended speech (like a familiar voice pattern) do get through the filter. However, the fact that you catch your name specifically hints that some meaning leaks through too, which is a challenge for strict early selection.
• Semantic priming studies support late selection. These experiments show that words presented in an unattended channel can influence responses to words in the attended channel, suggesting unattended stimuli are processed for meaning even when you're not aware of them.

Kahneman's Capacity Model

Rather than asking where the filter is, Daniel Kahneman asked a different question: how much attention do you have to work with? His capacity model (1973) treats attention not as a filter but as a limited pool of mental energy that gets distributed across tasks.

The model has several key components:

1. Central capacity pool. You have a finite amount of cognitive resources available at any moment. Every task you perform draws from this shared pool.

2. Allocation policy. Your brain distributes resources based on two factors:

   • Enduring dispositions are long-term influences like habits, overlearned skills, and reflexive responses (e.g., you automatically orient toward a loud noise).
   • Momentary intentions are your current goals and instructions (e.g., "pay attention to the speaker, not your phone").

3. Evaluation of demands. Your brain continuously assesses how much capacity each task requires and adjusts allocation accordingly.

4. Arousal level. Your overall arousal (how alert or activated you are) affects the total size of the resource pool. Higher arousal generally means more available capacity, up to a point.

This model explains why some task combinations are easy and others are nearly impossible. Walking and talking work fine together because walking is automatic and demands few resources (parallel processing). Reading a textbook while listening to a podcast on a different topic fails because both tasks compete for the same limited language-processing resources (serial processing). Task difficulty and your motivation also shift how much capacity is available and where it goes.

Implications of Attentional Theories

These theories aren't just abstract. They explain real phenomena and have practical applications:

• Change blindness occurs when you fail to notice even large visual changes in a scene, illustrating how much information goes unprocessed outside the focus of attention.
• Multitasking costs are predicted by capacity models. Texting while driving degrades performance on both tasks because they exceed available resources.
• Selective perception and bias show how attention shapes your subjective experience. You tend to notice information that confirms what you already believe (confirmation bias), partly because attention is directed toward expectation-consistent stimuli.
• Cognitive disorders like ADHD involve disruptions to attentional control, and these theories help frame what specifically is going wrong in terms of filtering or resource allocation.
• Interface design draws on attentional research to minimize distractions and highlight critical information, since designers know users have limited attentional capacity.