Lab Talk

Attention

Many Ways to Measure Attention

Attention is a multifaceted, multisensory cognitive phenomenon that can be studied in many ways.  Here we describe a few methods used in the lab to measure its different aspects.

Attention is a fundamental aspect of cognitive function.  Yet it is not a single thing but rather a multifaceted concept that includes dividing, switching, selecting, orienting, searching and sustaining. To explore these many facets of attention, numerous methodological approaches have evolved.  Here we outline 5 common types of tasks which are used to measure attention in the lab that can also be combined with neuroimaging techniques such as EEG.

  1. Tasks for measuring attentional orienting.

One of the most classic attention tasks is the spatial cueing task, sometimes called the “Posner” task. It measures how fast people can orient or shift their attention to particular locations in space in response to a cue.  The task involves the participant looking at a screen with a square in the center to the right or left of which an image will occur.  An arrow cues the participant which direction the image will occur.  The participant must press a key as soon as the image appears in the cued location.  Essentially this task measures how quickly we shift attention from the cue to the target when alerted. Often in some trials an invalid cue is presented (i.e. pointing the wrong direction) or the cue is omitted to see how this influences the reaction time.

Deficits in performance occur in some clinical disorders (e.g. ADHD) and become more evident with age. Multiple variants of the task exist using EEG. For example, studies from John Foxe’s lab have explored the role of occipital alpha activity in the anticipatory phase between cue and target in both the visual and auditory domain, whilst studies from Anna Nobre’s lab have demonstrated the role of alpha activity when participants orient their attention within a remembered spatial array (e.g. stored in their working memory).

Schematic of the Posner spatial cueing task. The top pane shows an example of endogenous trials where the cue is presented centrally, whilst the bottom panel shows  exogenous trials  where the cue is presented peripherally. Valid trials are when the target appears at the cued location, whilst during invalid trials the target appears at the non-cued location.

 

 

2. Tasks for measuring attentional control.

The attention network test (ANT), developed by Jin Fan (task can be downloaded here for research purposes), is a more recent paradigm that combines Posner’s cued reaction time task and Eriksen’s flanker task to measure three types of attentional control – alerting, orienting and executive (or decision making).  In addition to the standard Posner task, the test involves presentation of the cue arrow presentation flanked by either congruent (same direction) or incongruent arrows (opposite direction) which requires the participant to use more sophisticated (executive) decision making to determine which way the arrow is pointing. This 3-in-1 approach has been used to assess attentional performance in a wide number of clinical populations including autism, schizophrenia, anxiety and ADHD, and in combination with EEG to demonstrate age-related decline and other brain changes.

 

Schematic of the ANT. A fixation cross appears in the center of the screen all of the time. In each trial, depending on the cue condition (no cue, center cue, or spatial cue), a cue may appear for 200 ms. After a variable duration (300–1450 ms), the target (the center arrow) and flankers of two left and two right arrows (congruent or incongruent flankers) are presented. The participant makes a response to the target’s direction within a time window of 2000 ms. The target and flankers disappear after the response is made. The target and post-target fixation period lasts for a variable duration (3000–4200 ms). From Fan et al 2007.

 

3. Tasks for measuring sustained attention and vigilance.

There are a number of different vigilance tasks where participants have to sustain their attention over time, typically towards a continuous sequence of stimuli. These includes tasks such as the  sustained attention to response task (SART) and Conners continuous performance test (CPT), where participants are presented a continuous series of letters or images in a fixed or random fashion and have to press a spacebar whenever they see an X.  This requires both being alert to the X and withholding response to other letters.  The results have been shown to be sensitive to various clinical disorders (e.g. ADHD) and experimental interventions (e.g. caffeine administration). Other tasks which require sustained attention include the rapid serial visual presentation (RSVP) task, the psychomotor vigilance task (PVT) and the oddball task, where participants have to respond to a rare or deviant target. For example, the  rapid serial visual presentation (RSVP) task can be adapted to measure the attentional blink phenomenon when two targets are presented in close succession.

Schematic of the Sustained Attention to Response Task (SART), demonstrating the sequence of events and timings for the SART. Figure depicts (A) a go trial (requiring a response to the presentation of the go-digit 1), and (B) a no-go trial (requiring the withholding of a response to the no-go digit 3). In the Fixed version of the SART, the digits 1–9 are presented within a fixed sequence that is repeated 25 times. In the Random version of the SART, the digits are presented in a pseudo-random order. All participants respond on the response cue. Taken from Johnson et al 2007

 

4. Task for measuring visual search.

 Visual search uses a range of stimulus types from single letters through to complex scenes (see here for a large visual search dataset from Jeremy Wolfe lab). In both cases, participants have to search the array or scene for the target stimulus, keeping in mind what they are looking for and ignoring irrelevant distractors. There are various theories about how visual search best operates from the perspective of the brain and EEG studies can help to provide insights into the factors which influence searching performance.

Example of visual search tasks using simple stimuli requires you to find the odd one out.

(from http://search.bwh.harvard.edu/new/index.html)

 

  1. Measuring attentional biases to emotion.

 The dot probe task allows researchers to measure the way attention is biased by emotional stimuli (e.g. by threatening information). The classical version of the task is a covert attention task with words or faces (a word version of the task can be downloaded here for research purposes), and allows the researcher to measure how rapidly the emotionally charged information “grabs” the participant’s attention. Strong biases to threatening or negative stimuli are often observed in patients with anxiety, however the reliability of the task is not always consistent and is a reminder that just because a task is used frequently, doesn’t necessarily mean it is always the best task to use.

A schematic of the dot probe task. On the left is a congruent trial where the dots are presented in the location of the emotional stimulus, whilst on the right is an incongruent trial where the dots are presented in the location of the neutral stimulus.

In summary, attention is a multifaceted phenomenon and there are a range of tasks to measure these various facets.  While we have presented a few that are primarily visual in nature, attention also extends to other sensory modalities and may indeed be different across modalities.  A person may, for example, find directing visual attention easy but auditory attention more difficult.  Furthermore, not all methods are discussed here – there are a host of others as well, for example, relating to divided attention (e.g. between two modalities), attentional interference (e.g. stroop tasks) or attentional flexibility.  Lastly,  we point out that these tasks are simple in nature and how they relate to more complex paradigms is not clear.  Indeed, there is no single test to truly assess the strengths and weaknesses in someone’s attentional functioning.  Rather it would require a battery of tasks of increasing complexity that integrate different facets of attention and sensory modalities.

One thought on “Many Ways to Measure Attention

  1. These are all traditional and well-documented experimental paradigms. But I would argue that the question of construct validity – whether what the tasks aim to measure actually maps on to a parameter of individual differences that is meaningful in the real world – is relatively unproven. See here https://www.sites.google.com/site/samwass/blog-for-researchers/untitledpost for a discussion of this – and see here https://www.sites.google.com/site/samwass/blog-for-researchers/onyerkes-dodsonandnaturalisticattention for a discussion of more naturalistic, ecologically valid approaches to measuring attention…

Leave a Reply