Attention is a cognitive process that allows to select some environmental stimuli, ignoring others. From an evolutionary point of view, it is an extremely useful mechanism for human survival because it allows us to organize the information coming from the external environment, which is constantly changing, and to regulate mental processes accordingly.
Attention refers to the process by which organisms select a subset of available information upon which to focus for enhanced processing (often in a signal-to-noise-ratio sense) and integration. Attention is usually considered to have at least three aspects: orienting, filtering, and searching, and can either be focused upon a single information source or divided among several. Each of these aspects has specific properties that are discussed briefly below. Attention and consciousness are closely related although the two concepts can be both conceptually and empirically distinguished.
In general, in psychology we can define attention as the set of selection processes that the brain puts in place against the stimuli that come from the outside world through the sense organs. A metaphor often used is that of the filter, which allows only relevant stimuli to pass.
Already at the dawn of psychological sciences William James observed that man cannot be aware of everything, in the face of a very high amount of data and sensory information. In fact, the human cognitive system is a limited system, that is, it has a limited amount of resources for information processing (Broadbent, 1958). In this sense, attention can be traced back to the set of mechanisms and processes that allow us to focus our mental resources on some stimuli or information at the expense of others, determining what we are aware of at a given time.
The psychology of attention aims to study attentional processes through the use of specific experimental paradigms, techniques and tools. In addition to neuroimaging studies, which aim to identify brain activity in relation to attentional processes, there are also traditional techniques still used in the literature. For example, it is possible to study the behavior of subjects called to perform certain attentional tasks in the laboratory by measuring their reaction time (TR). Reaction time is the amount of time that elapses between the presentation of a stimulus and the emission of a response. The longer the time between stimulus and response, the more processing is required. Some experimental paradigms are instead aimed at investigating the relationship between attention, perception of objects or events, and consciousness.
When talking about attention and effective performance, it is important to refer to the theory of arousal, understood as a global state of activation of the individual that can vary from sleep to widespread arousal. Attention is a function that correlates with the level of global activation. The relationship between activation level, alertness, and efficiency in performing tasks or performances is represented by an inverted U-shaped curve. At low levels of activation, the individual is easily distracted (with negative effects in terms of task efficiency), while, likewise, at excessive levels of activation, anxiety negatively impacts performance (Yerkes and Dodson Theory). According to Yerkes and Dodson’s theory, both too basic and too high levels of activation would also increase distractibility and a deterioration in performance.
The simplest way to select among several stimulus inputs is to orient our sensory receptors toward one set of stimuli and away from another. Seeing and hearing are not usually passive, but rather, we actively look or listen in order to see and hear.
A prototype for orienting is the response of a dog or a cat to a sudden sound. The animal rapidly adjusts its sense organs, by pricking its ears and turning its eyes, head and/or body, so it can optimally pick up information about the event. Responses such as flicking the eye in the direction of a sound or peripheral movement, as well as accompanying postural adjustments, skin conductance changes, pupil dilation, decrease in heart rate, a pause in breathing, and constriction of the peripheral blood vessels, occur automatically and are collectively referred to as the orienting reflex.
The most effective orienting stimuli are loud sounds, suddenly-appearing bright lights, changes in contours, or movements in the peripheral visual field that are not regular, predictable occurrences. It is as though we had an internal ‘model’ of the immediate world of stimuli around us.
When we notice a departure of stimulus input from that model, we reflexively orient to that stimulus in order to update that model as quickly as possible (Sokolov, 1975). If the same stimulus occurs repeatedly it becomes an expected part of our model of the world, and our orienting reflex toward it becomes weaker, even if the stimulus is quite strong. With a change in the nature of the stimulus, however, the reflex recovers to full strength.
The overt orienting response to sudden changes in the environment is usually accompanied by another, unseen orienting response, the fixing of attention on the event or object that elicited the reflex. This unseen attentional orienting is called covert orienting. The combination of overt and covert orienting to an event usually results in enhanced perception of that event, including faster identification and awareness of its significance: for example, we can move our eyes to an object (overt attention) or shift attention to it without moving the eyes (covert attention).
Although this hidden orienting of attention usually occurs in association with overt orienting, whether reflexive or voluntary, it is possible to covertly attend to an event or stimulus without making any overt sign that we are doing so (e.g., Helmholtz, 1867/1925; Posner, 1980; Wright & Ward, 2008).
Thus, covert attention orienting is usually studied separately from overt orienting behaviors, although the two are surely closely related. The premotor theory of attention (e.g., Rizzolatti et al, 1994) proposes that these attention systems are supported by the same neuronal mechanisms, although a detailed examination of the behavioral and physiological data indicates that whereas the two probably share some neural mechanisms they are not identical (see discussions by Corbetta & Shulman, 2002; Wright & Ward, 2008).
Stimulus-driven, exogenous, orienting
Just as in the orienting reflex, abrupt-onset or intense stimuli can cause covert orienting, that is, they capture attention. For example, abruptly-appearing letters on a computer monitor capture attention and are responded to faster than are gradually-appearing letters (Jonides & Yantis, 1988; Yantis & Jonides, 1984). If such an abrupt-onset stimulus (a direct cue) appears about 100 msec before another stimulus (a target) in the same spatial location, the latter is processed faster and more accurately than if it had appeared in another location (e.g., Müller & Humphreys, 1991), presumably because attention was attracted reflexively to the spatial location of the direct cue.
Attention that is captured reflexively in this way is said to be oriented exogenously in a stimulus-driven manner. Attention oriented in this stimulus-driven manner doesn’t remain oriented to the attracting location for long, however, moving to another location after 100-200 msec or so unless the attracting stimulus requires enhanced processing or signals a high probability of a target occurring at that location.
Attention also is drawn to the abrupt appearance of a new perceptual object even if its appearance is not accompanied by a luminance change but not to a luminance change that is not associated with the appearance of a new object (Yantis & Hillstrom, 1994). Thus, attention is drawn both to a new perceptual object and to its location (Egeth & Yantis, 1997). Exogenous attention capture also happens in hearing and in touch.
Moreover, a direct cue in one of these modalities, say a sound, can orient attention to a location so that when a target in another modality, say a visual pattern, occurs there it too is processed more rapidly and accurately (Wright & Ward, 2008).
Part of the cue effect in stimulus-driven orienting appears to be caused by residual sensory activity from the direct cue itself, which also dissipates within 100-200 msec (Wright & Ward, 2008). Indeed, simultaneous presentation of up to 4 direct cues in the same display can cause cue effects at all of their locations (Wright & Richard, 2003).
Nonetheless, the cue effect from a single direct cue display is larger than those from multiple cue displays, indicating that cue effects from a single direct cue arise both from orienting attention and from residual sensory activation, whereas those from multiple cue displays arise solely from residual sensory activation (Wright & Richard, 2003).
Neurophysiology of stimulus-driven orienting
Reflexive attention capture by abrupt or intense stimuli is implemented by a network of brain areas that includes the superior colliculus, the pulvinar nucleus of the thalamus (both subcortical), and the posterior parietal cortex, as well as areas in the frontal cortex and of course the various sensory cortices. Much research suggests that an early model, in which the posterior parietal cortex disengages attention from a current target/location, the superior colliculus shifts it to a new target/location, and the pulvinar nucleus engages attention at that new locus, is roughly correct (Wright & Ward, 2008).
Recent imaging research has revealed a more specific picture of the cortical parts of this network, which include the temporal-parietal junction and the ventral-frontal cortex, specifically the inferior frontal gyrus and the middle frontal gyrus, mainly on the right side of the brain (Corbetta & Shulman, 2002).
Goal-driven, or endogenous, orienting
In addition to reflexive, stimulus-driven orienting, we can also orient attention to a location in space or to an object voluntarily (endogenously) in a goal-driven manner, often based on a cue that tells us where to look or to listen, for example a loudspeaker announcement that passengers from a particular flight will disembark at a particular gate in the airport. Information about where or what to look at or to listen for an expectancy about an environmental event, and we often prepare for the event by orienting attention to the location and to the time of the expected event (LaBerge, 1995). This advance, goal-driven alignment of attention enhances processing of the event when it does happen (e.g., Posner, 1980).
Costs and benefits of symbolic cues
Goal-driven orienting is usually studied in experiments by presenting a predictive symbolic cue (such as an arrow pointing to a possible target location) as to where in space a target stimulus will occur, the so-called Posner paradigm (e.g. Posner 1980). Usually (say on 80% of arrow-cue trials, which may represent 80% or more of total trials) the target does occur there (called valid-cue trials), giving the subject an incentive to focus attention there in advance of the target’s appearance. Sometimes (on 20% of arrow-cue trials when 80% are valid) the target occurs at a location other than the cued location (called invalid-cue trials).
Finally, in some implementations a third group of trials (neutral-cue trials; often 20% of total trials or less) a neutral (as to target location) cue is presented. On these trials targets occur at random at possible locations. Under these conditions, targets on valid-cue trials are responded to faster than are targets on neutral trials (a benefit of orienting to the target location), whereas targets on invalid-cue trials are responded to more slowly than are targets on neutral trials (a cost of orienting to the wrong location).
Such goal-driven orienting of attention is slower than is stimulus-driven orienting, taking usually about 300 msec to reach full effectiveness (Shepard & Müller, 1989). Moreover, goal-driven attention can be sustained at a location for quite long periods, even several minutes, whereas stimulus-driven attention is usually transient unless goal-driven attention is invoked by a predictive direct cue or an interesting target stimulus. Goal-driven orienting is not automatic, and can be interrupted by an attention-capturing stimulus (Müller & Rabbitt, 1989), although it can also be sustained even in the presence of such a stimulus if the predictive value of the symbolic cue is high enough (Yantis & Jonides, 1990).
Neurophysiology of goal-driven orienting
A specific network of brain areas subserves goal-driven attention orienting. The cortical parts are called the dorsoparietal network (Corbetta & Shulman, 2002). It involves frontal regions, specifically the frontal eye fields (FEF), which seem also to be involved in voluntary orienting to auditory and tactile events (e.g., Shomstein & Yantis, 2004), parietal regions, specifically the intraparietal sulcus (IPS), as well as the relevant sensory cortices and subcortical areas (thalamus and superior colliculus).
Moreover, this network interacts with the ventroparietal network (shown in red), probably through connections between the TPJ and the IPS (shown in blue). It is possible that the interactions between the brain regions of the dorsoparietal and ventroparietal networks are mediated by synchronization of their activites at various frequencies, including especially those in the gamma (30-70 Hz) and in the alpha (8-14 Hz) ranges (e.g., Doesburg, et al, 2008; see also Varela et al 2001; Ward, 2003).
Inhibition of return
When attention has been drawn to a particular location or perceptual object and then moves away to another location or object, it seems to be inhibited from returning to the original location or object for a period of up to 2 sec (e.g., Posner & Cohen, 1984; Tipper, Driver, & Weaver, 1991). This is called inhibition of return (IOR).
IOR is supposed to promote the search for informative objects or locations, and it occurs within and across visual, auditory and tactile modalities, just as attention does (e.g., Klein, 2000; Spence & Driver, 1998; Ward, 1994), and in infants as young as 6 months of age (Rothbart, Posner, & Boylan, 1990). Although it is possible that IOR arises from inhibition attached to some motor process such as eye movements or manual responses (see Wright & Ward, 2008 for a review of premotor theory in this regard), recent evidence indicates that it is also possible that its neural expression arises during perceptual or cognitive processing of stimuli at previously attended locations (e.g., Prime & Ward, 2006).
When attention is oriented to a particular locus in space it is called focal or spatial attention. If it is oriented to a particular object, whether visual, auditory, somatosenory, smell, or taste, it is called object-oriented attention. But whether to a place or to an object, the extent of the attentional focus can be controlled in a goal-driven or a stimulus-driven manner (see Yantis & Serences, 2003, for a review of space-based and object-based attention and their similar cortical mechanisms).
The greater the extent over which attention is spread, the less efficient is the processing of information within that area (e.g., Laberge & Brown, 1989). Moreover, the farther a stimulus is away from the center of an attended region, the less efficient is processing (e.g., Eriksen & St. James, 1986). This latter effect is sometimes called the attention gradient.
Attention acts as a filter, extracting more information from attended stimuli and suppressing information extraction from unattended stimuli. The suppression can be so great as to cause what has been called inattentional blindness, which occurs for all modalities.
The cocktail party phenomenon
At a noisy cocktail party (or any other kind) attentional filtering is rampant. People listen to one conversation, line of music, etc. and filter out the rest as ‘noise.’ Cherry (1953) used shadowing to study this phenomenon in auditory attention. In this technique, an observer must repeat aloud (shadow) one of two continuous speech streams. Close shadowing results in loss of most information in the non-shadowed stream except sometimes for powerful direct cues like one’s own name or a loud noise, which can cause orienting to that stream. The non-shadowed information does get to short-term memory, however, and can be recalled if the shadowing is interrupted.
Inattentional and change blindness
An analog of auditory shadowing was used to study visual filtering (Neisser & Becklin, 1975). Subjects shadowed one of two overlapping video programs with similar results: most of the information in the non-shadowed stream was not noticed or remembered. Like auditory filtering, visual filtering allows little of the filtered out information to make a lasting impression.
Modern replications of this result have produced even more dramatic, and counter-intuitive, results, for example a failure to notice a person in a gorilla suit thumping his chest and cavorting around during a shadowed basketball game (Simons & Chabris, 1999). Even quite dramatic stimuli in simple displays can pass by unnoticed if attention is directed elsewhere. This has been called inattentional blindness (Mack & Rock, 1998). A related phenomenon is change blindness (e.g., Rensink, 2002), in which a scene and the same scene with a change in it are presented for brief periods separated by a blank interval. Until attention alights on the changed element it is not reportable.
Neurophysiology of filtering
Attentional filtering seems to be accomplished by the activity of the pulvinar nucleus of the thalamus under the instructions from other, probably frontal, cortical areas such as the frontal eye fields (FEF; LaBerge, 1995). The pulvinar nucleus is also supposed to be the subcortical area responsible for engaging attention, and it shows elevated activity when attention must be used to filter out distracting stimuli (e.g., Corbetta, et al., 1991; Laberge & Buchsbaum, 1990).
The effect of attending to a specific stimulus on neural activity is manifest quite early in sensory processing and includes both enhanced response to the attended stimulus and inhibited response to other, unattended stimuli. For example, in vision, Moran and Desimone (1985) showed that responses in monkeys’ visual area V4 to the same stimulus depended greatly on whether it was attended or ignored. Appropriate attended stimuli in the receptive field of a V4 neuron received a vigorous response that was dramatically reduced when that same stimulus had to be ignored.
Similarly, attending to a specific feature of objects in the visual field generates more vigorous responses from the neurons in monkey V4 tuned to those features than it does when attention is directed to other features of the same objects (see review by Maunsell and Treue, 2006).
Moreover, gamma-band synchronization of neural responses from different neurons in V4 to the same stimulus is increased by selective attention, whereas synchronization in the alpha-band is decreased, possibly serving to amplify the signal generated by the attended stimuli relative to that generated by distractors (Fries et al, 2001). A good review of these mechanisms that is still current is Kastener and Ungerleider (2000).
The term divided attention (or split attention) refers to an individual’s ability to focus attention on more than one stimulus or event at the same time. In fact, in everyday life it is a widespread psychological phenomenon and it allows us to perform several activities at once, such as listening to a radio program while driving a car.
In the field of experimental psychology, researchers have mainly used the dual-task paradigm to study the phenomenon of divided attention. The dual-task paradigm implies that the subject is engaged in two experimental tasks simultaneously, calling into play different abilities or having different levels of difficulty.
In everyday or experimental situations, when we are faced with a double task, two types of cognitive processes come into play: control processes and automatic processes. The former occur under conscious control, are slower and require a higher cognitive load, while the automatic processes are faster and unconscious. Generally, the task on which we are more experienced may require more use of automatic processes (already established learning, e.g. cycling) without interference between the two tasks. In contrast, in the case where two tasks require conscious, serial control processes, we face competition for limited resources and a deterioration in performance.
Regarding divided attention, on the other hand, so-called capacity theories support the divisibility of cognitive resources between different tasks performed simultaneously and the partial use of attentional resources for different tasks. For example, Kahneman’s model integrating the structural theories with those of capacity, firstly recognizes the inherent limit in the resources of the individual for the execution of mental tasks. Second, according to Kahneman’s model there would be a gradual increase in mobilized resources as a function of demands in the case of a dual task, although this gradual increase reaches a threshold limit. At this threshold, demands exceed the individual’s resources and interference between tasks is observed.
Dividing attention between two (or more) sources is very difficult. For instance, people can’t easily listen to two simultaneous audio streams or view two overlapping videos while detecting target events in each, especially when the two sources are spatially separated. Sometimes two aspects of a single object can be attended to successfully, but if the two aspects characterize two spatially separated objects performance is worse in divided attention conditions (Bonnel & Prinzmetal, 1998).
It is also easier to divide attention between information streams in two different sensory modalities, such as vision and hearing, but if the task is more difficult than simply detecting occasional stimuli in those channels performance is still worse than if attending to only one channel (e.g., Bonnel & Hafter, 1998).
When the task is more difficult, only if the task in one modality, say typing by an expert typist, can be performed automatically can attention be divided without a performance decrement, and then only if the response modalities are similarly different (e.g., expert typist typing a text – visual/manual while making a verbal response whenever they hear their name in an auditory channel – auditory/verbal).
When we know what we are looking for but don’t know where to find it we must search for it. Attention is importantly involved in this search, and search experiments have yielded much information about its mechanisms.
Easy versus difficult search
When observers are presented with a field of items to search for a particular target, they can perform the search very quickly, and in roughly the same time regardless of the number of non-target items, if the target differs from the non-targets in a single feature; this is called easy/pop-out/parallel search. If the searcher must detect a conjunction of features, however, the search is slow, and the time to find the target increases linearly with the number of non-targets; this is called difficult/serial search.
Easy search could be accomplished by simply detecting the presence of activation in a particular one of the feature maps generated by sensory processing (feature maps are sheets of visual cortex in which activity of particular neurons signals the presence of a particular feature, such as a slanted line, on the retina). This can be done without orienting attention to any one specific item in the field. On the other hand difficult search seems to require focusing attention on each item in turn, which slows the search and makes the overall search time dependent on the number of non-target items (e.g., Woodman & Luck, 1999).
Feature integration theory (e.g., Treisman & Gelade, 1980) explains these data by assuming that the binding of features together into a perceptual object requires focal attention to a particular locus. This is not the full story, however, because conjunctions of simple features sometimes also result in a very rapid search (e.g., Wolfe et al, 1989), and attention can be oriented sequentially to items in easy searches and to individual features in conjunction searches (Kim & Cave, 1995).
Automatic versus controlled search
Much practice on a difficult search task that always demands the same response for a given stimulus gradually turns the process from controlled search, in which search time is a function of the number of items in the search set, to automatic search, in which search time is roughly independent of the number of items in the search set (e.g., Schneider & Shiffrin, 1977; Shiffrin & Schneider, 1977). Search functions for controlled and automatic search resemble those for easy and difficult search.
A common explanation for the process of automatization is that attention is slowly withdrawn from task control with increasing practice, until the process requires only a minimal amount and is said to be automatic. In automatic search responses are ballistic, so that they are difficult to inhibit, and also aren’t remembered very well. Many attention lapses in daily life can result from such automatic responses (e.g., Reason, 1984). On the other hand, automatic processing does enable better divided attention performance.
Theories of attention
Each of the major aspects of attention, orienting, filtering, and search, has spawned numerous theories, both at the psychological and at the neurological level. Early theories, focused on filtering, claimed that information processing in the brain was structurally limited, with an early filter based on physical characteristics such as location and spectral content, that allowed only a selected few stimuli past it (e.g., Broadbent, 1958).
Demonstrations that at least some processing is done even on the rejected channels led to the rejection of these early selection theories in favor of late selection theories that hypothesized that all sensory information receives preliminary analysis. The processing bottleneck instead occurs just before entry into longer lasting memory (e.g., Deutsch & Deutsch, 1963). There is physiological evidence favoring both approaches and probably both occur in different circumstances (e.g., Pashler, 1996).
Failures of divided attention have led to the idea that attention is a limited resource that, being demanded by one task, is unavailable for another (e.g., Kahneman, 1973). The demonstration that time sharing among perceptual or cognitive tasks is possible, especially if one is overlearned, or automatic, on the other hand has led to the idea that there are multiple attentional resources, and that these can be divided among tasks providing there are no conflicts (e.g., Wickens, 1984).
According to this approach, central resources (such as encoding, comparing, remembering) interact with spatial and verbal codes, the sensory modalities, and the response systems to constrain performance. This approach has been criticized as being too flexible (e.g., Navon, 1984), although the idea of limited attentional resources is still used widely.
Another class of theories attempts to capture the mechanisms of attention orienting. Some of these have already been mentioned earlier. The approaches of Corbetta and Shulman (e.g., 2002), discussed above, Laberge (e.g., 1995), and of Shipp (e.g., 2004) are among the most general and useful. Many such theories emphasize the concept of salience, that is that attention orienting is to the most salient of available locations or objects, with salience being a combination of bottom-up and top-down contributions.
Shipp (2004) discusses several salience models and combines them all in a theory of the physiology of the orienting system, with a salience map in the pulvinar nucleus of the thalamus combining inputs from other such maps throughout the brain. In this theory, IOR is explained by a decrement in the salience of a location or object that has been inspected recently.
Finally, there are other models that attempt to capture particular aspects of attention in computational or mathematical frameworks. The episodic theory of Sperling and Weichselgartner (1995) is particularly general, and accommodates a variety of possible mechanisms. In this approach, general transition functions of time are assumed for attention movement and onset and offset at particular locations. These temporal functions refer to an attention spotlight that can have a variety of characteristics such as extent and intensity.
Thus the theory can accommodate a variety of results from studies of all of the aspects discussed earlier. More recently, Taylor (e.g., Taylor & Rodgers, 2002, CODAM Model) has proposed a control model of attention movement that is expressed as a neural network and that is neurophysiologically realistic. There are also theories of attentional oscillations and the entrainment of attentional focus to rhythmical events such as music (e.g., Large & Jones, 1999). These dynamical theories are also mathematical, since the description of oscillators is. None of these theories has yet gained universal acceptance. Follow the links below to reach discussions of several specific theories.
The Control of Attention
Attention can be directed voluntarily (or endogenously) or automatically.
An example of voluntarily directed attention is found in Posner’s classic experiment. In this experiment the “spatial cueing paradigm” is used to study the shift in attention. The subject stands in front of a monitor and is asked to keep his gaze fixed on a fixation point (a cross on the screen, for example). Next to the cross there are also two squares and, during the experiment, a target stimulus appears at certain times within these squares. The subject’s task is to detect the appearance of the target as quickly as possible. In addition, an arrow appears just before the square, suggesting with a high probability the position of the target’s appearance.
By measuring the subject’s reaction time, it was seen that individuals responded faster when the arrow correctly indicated the target’s position of appearance. In this sense, subjects were able to preemptively shift their attention to the indicated location and this allowed for faster information processing. In this paradigm of spatial suggestion, attention is directed voluntarily, as the subject directs attention into the square indicated by the arrow, being aware that with a reasonable probability the arrow correctly precedes the appearance of the target.
Attention can also be directed automatically (or exogenously), that is, independently of the subject’s will. For example, this type of automatic attention orientation is observed when a new and unexpected peripheral light signal appears. Automatic can be defined as an orientation of attention that is independent of cognitive load and is resistant to suppression, acting almost as a reflex.
Beyond laboratory experiments, in our daily life it often happens that the components of voluntary and automatic orientation of attention are coexistent. For example, if our goal is to look for something, it often happens that our attention is distracted by the presence of another object. In general, it is believed that attention can be captured automatically by events, stimuli and information irrelevant to the subject’s purpose and task.
Selective attention refers to the ability to focus on the target stimulus, the object of interest, and to process in a privileged way the information relevant to the achievement of a specific purpose. The information to which attention is paid is selected and processed more efficiently, has access to consciousness and guides the response.
Faced with a complex environment full of stimuli, the individual to achieve a goal and / or implement a behavior must be able to select some of these stimuli-objects and neglect others. Through selective attention and spatial attention the focus of attention is directed to a delimited portion of the space and some stimuli that fall within the focus of attention are considered as relevant and reach the level of consciousness.
A typical example is the phenomenon of the cocktail party in which you can pay attention to a single conversation despite there are many others that could interfere: although the sound emissions from different guests are captured by the acoustic receptors, the individual through selective attention selects and analyzes only those coming from some interlocutors.
Several theoretical contributions have attempted to understand and explain in which moment of the process of information processing, attention intervenes and the selection of information to be processed takes place. A first theoretical approach refers to the early selection of information, while other theories advocate a late selection of information. Both of these approaches share the idea that attention acts by means of a sort of filter that allows the passage and processing of only a limited amount of information. The difference lies in the moment in which this filter intervenes considering the whole process of elaboration, that is, at an earlier or later stage.
According to the theories of early selection, attention acts as a filter that excludes from processing most of the information that comes from outside and the selection is already implemented at the level of sensory input. An example of a theory of early selection of information is Broadbent’s filter theory which refers to the dichotic listening paradigm. This paradigm requires the subject to pay attention only to messages that are transmitted to one ear, ignoring information transmitted to the other ear. According to this theory, there would be an initial phase of information processing during which all stimuli are rapidly analyzed in parallel by the sensory system (S) simultaneously and stored for a short period. This initial processing would be followed by a more advanced phase of serial processing by the perceptual system (P). A filter, placed between the S system and the P system, selects stimuli that can have access to the more advanced levels of processing.
A similar theory is Treisman’s attenuated filter theory. Building on Broadbent’s classic filter theory, the author revises the very concept of attentional filtering from an attenuation perspective: the filter would not completely eliminate the processing of irrelevant information, but would attenuate their processing and keep them subthreshold. Such subthreshold stimuli, under certain conditions, could be reactivated and reprocessed in a more complex and conscious way by the subject.
Thus, according to the early selection approach, information selection occurs before the semantic content is processed, with a progressive decay of the information that is not selected because it is considered irrelevant upstream. According to late selection theories, on the other hand, the processing of irrelevant information is nevertheless complete, not decaying a priori.
For example, Deutsch and Deutsch’s (1963) theory is based on the assumption that all information and stimuli, of varying relevance, are processed at the semantic level. Selective attention comes into play only when a response must be selected to be issued. In this sense, selective attention allows to control the access of information to consciousness.
One of the classic experiments on selective attention is the Stroop test. During the Stroop experiment the subject is shown words written in different colors. The task is to say out loud the color of the ink with which the word is written. Thus, the color is the relevant information for performing the task, while the meaning of the word (which is not to be read) is the irrelevant information. The stimuli presented in Stroop’s experiment can be neutral, congruent, and incongruent. Neutral is when only text or only color is displayed. Whereas, there is congruence when the word “red” is written in red, and incongruence when the word “red” is written in green. Recall that the required response is the name of the color, i.e., red in the former case and green in the latter.
Stroop (1935) noted that participants subjected to the naming task exhibited slower response times if the color of the ink was different from the meaning of the written word, despite being instructed to disregard the meaning of the word. The Stroop effect, then, is to produce a response having slower latency in the case of the incongruent condition and faster latency in the case of the congruent condition. The purpose of the Stroop experiment is to create a cognitive and semantic interference: in this case, for example, the mind tends to mechanically read the meaning of the word (for example it reads the word “red” and thinks of the color “red”, but the ink used is a different color). For this reason, the Stroop test is a well-established experimental procedure for studying selective attention.
As supported by the theories of late selection, the individual would put in place a kind of processing even to those information or stimuli to which it does not pay direct and “conscious” attention in a voluntary way.
With the advent of cognitivism, the first studies that related attention and consciousness focused on the so-called subliminal perception. With the term “subliminal perception” we refer in psychology to those phenomena in which a stimulus influences behavior even if it is “unconsciously grasped” by the subject, for example because it is presented too fast.
In studies using the priming paradigm, a stimulus called “prime” is presented first and, immediately afterwards, a second stimulus called “target” is presented which implies the emission of a response by the subject. Some studies related to the speed of linguistic comprehension have provided empirical support for the theories of delayed selection: the subject would understand with less reaction time words semantically related to the prime. Conversely, when the target stimulus to be recognized is not semantically related to the prime, reaction times increase. In this sense, the explanatory hypothesis is that the representations of irrelevant information would be initially activated and semantically processed and, only later, inhibited because they are irrelevant.
Attention and consciousness
Attention is closely related to consciousness. Both are integrative and yet also selective. Inattentional blindness seems to indicate that items not in attention are not consciously perceived. Indeed so close is this relationship that Taylor’s CODAM model of attention is asserted also to provide, through the corollary discharge of the attention control signals, two important aspects of consciousness, viz. the sense of ownership of the conscious experience and its immunity to error though misidentification.
Nonetheless, the two concepts are differentiable. Attention is usually conceptualized as the boosting of signal-to-noise ratio both through inhibition of processing of unattended stimuli and through enhanced processing of attended stimuli. Consciousness refers primarily to phenomenal experience itself, and secondarily to aspects of that experience such as its wholeness, its feeling of self-ownership (first-person ontogeny), the ability to report its contents verbally or in other ways, and the awareness of being conscious (metaconsciousness).
There are several models of the relationship between attention and primary, or phenomenal, consciousness, including that assumed by Taylor and most others in which cognitive material (sensations, perceptions, cognitions, memories, etc.) are either attended or unattended, with the attended items being experienced and also reportable, etc. One competing model, among others, has all cognitive material being either conscious or unconscious, with attention selecting some of the conscious material for enhanced processing, rendering it reportable (e.g., Lamme, 2003).
Lamme’s model assumes that all reentrant neural processing gives rise to conscious experience but that only experience selected by attention is reportable. This implies that there can be consciousness without attention but that only conscious material can be attended to. In contrast, Koch and Tsuchiya (2007) adduce evidence for a more complete dissociation between the two, that is, for attention both with and without consciousness, and consciousness both with and without attention. In this latter view, attention and consciousness are separate processes in the brain but have close links. As yet there is no definitive way to choose among these models.
Attention Deficit Hyperactivity Disorder (ADHD)
The basic clinical manifestations of ADHD are the difficulty in paying attention, the implementation of impulsive behaviors and/or the presence of a heightened level of motor activity.
We talk about ADHD with predominant inattention when the central problem of the child is precisely the attentional deficit. Selective attention and sustained attention appear to be the most impaired in this type of ADHD, but executive functions, especially planning and working memory, are also impaired. This discontinuity of attention compromises learning and does not allow the development of cognitive skills such as problem solving and behavioral strategies appropriate to establish satisfactory relationships with adults and peers.
We speak of ADHD with impulsivity and hyperactivity predominant, however, when the attentional function is mildly impaired, while the focus of the disorder lies in hyperkinetic behavior and lack of self-regulation. These deficits translate into disproportionate and inappropriate motor activation, excessive speech, difficulty in inhibiting responses and difficulty in respecting rules and turns.
- Attention by Lawrence M. Ward is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
- Bonnel A-M Hafter E. (1998) Divided attention between simultaneous auditory and visual signals. Percep. Psychophys. 60:179–190.
- Bonnel A-M Prinzmetal W. (1998) Dividing attention between the color and the shape of objects. Percep. Psychophys. 60:113-124.
- Broadbent DE. (1958) Perception and communication. New York: Pergamon Press.
- Cherry C. (1953) Some experiments on the recognition of speech with one and two ears. J. Acoust. Soc. Amer. 25:975–979.
- Corbetta M Miezin F Dobmeyer S Shulman G Petersen S. (1991) Selective and divided attention during visual discrimination of shape, color, and speed: Functional anatomy by positron emission tomography. J. Neurosci. 11:2383–2402.
- Corbetta M Shulman GL. (2002) Control of goal-directed and stimulus-driven attention in the brain. Nature Rev. Neurosci. 3:201–215.
- Deutsch JA Deutsch D. (1963) Attention: Some theoretical considerations. Psychol. Rev. 70:80—90.
- Doesburg SM Roggeveen AB Kitajo K Ward LM (2008) Large-scale gamma-band phase synchronization and selective attention. Cereb. Cortex 18:386-396.
- Egeth HE Yantis S (1997) Visual attention: Control, representation, and time course. Ann. Rev. Psychol. 48:269–297.
- Fries P Reynolds JH Rorie AE Desimone R (2001) Modulation of oscillatory neural synchrony by selective attention. Science 291:1560-1563.
- Helmholtz H von (1867/1925) Treatise on physiological optics (Trans. From 3rd German edition). New York: Dover.
- Jonides J Yantis S. (1988) Uniqueness of abrupt onset in capturing attention. Percep. Psychophys. 43:346–354.
- Kahneman D. (1973) Attention and Effort. Englewood Cliffs, NJ: Prentice-Hall.
- Kastner S Ungerleider LG (2000) Mechanisms of visual attention in the human cortex. Ann. Rev. Neurosci. 23:315-341.
- Klein RE. (2000) Inhibition of return. Trends Cog. Sci. 4:138-147.
- Koch C Tsuchiya N (2007) Attention and consciousness: Two distinct brain processes. Trends Cog. Sci. 11:16-22.
- LaBerge D (1995) Attentional processing: The brain’s art of mindfulness. Cambridge, MA: Harvard University Press.
- LaBerge D Buchsbaum MS (1990) Positron emission tomographic measurements of pulvinar activity during an attention task. J. Neurosci. 10:613–619.
- Lamme VAF. (2003). Why visual attention and awareness are different. Trends Cog. Sci. 7:12–18.
- Large EW Jones MR (1999) The dynamics of attending: How people track time-varying events. Psychol. Rev. 106:119—159.
- Mack A Rock I. (1998) Inattentional blindness. Cambridge, MA: MIT Press.
- Maunsell JHR Treue S (2006) Feature-based attention in visual cortex. Trends in Neurosci. 29:317-322.
- Moran J Desimone R (1985) Selective attention gates visual processing in the extrastriate cortex. Science 229:782-784.
- Müller HJ Humphreys GW. (1991) Luminance-increment detection: Capacity-limited or not? J. Exp. Psychol.: Hum. Percep. Perfor. 17:107–124.
- Navon D. (1984) Resources: A theoretical soup stone? Psychol. Rev. 91: 216–234
- Neisser U Becklin R. (1975) Selective looking: Attending to visually specified events. Cog. Psychol. 7: 480-494.
- Pashler H. (1996) The Psychology of Attention. Cambridge, MA: MIT Press.
- Posner MI. (1980) Orienting of attention. Quart. J. Exp. Psychol. 32:3–25.
- Posner MI Cohen Y. (1984) Components of visual attention. In H. Bouma DG Bouhuis (Eds.), Attention & Performance X (531–556). Hillsdale, N.J.: Erlbaum.
- Prime DJ Ward LM. (2006) Cortical expressions of inhibition of return. Brain Res. 1072:161–174.
- Reason J. (1984) Lapses of attention in everyday life. In R Parasuramen DR Davies (Eds) Varieties of Attention (515-549). Orlando: Academic.
- Rensink RA. (2002) Change detection. Ann. Rev. Psychol. 53:245–277.
- Rizzolatti G Riggio L Sheliga BM (1994) Space and selective attention. In C Umilta M Moscovitch (Eds) Attention and Performance XV (231-265). Cambridge, MA: MIT Press.
- Rothbart MK Posner MI Boylan A. (1990) Regulatory mechanisms in infant development. In JT Enns (Ed) The Development of Attention: Research and Theory (47-66). Amsterdam: Elsevier.
- Schneider W Shiffrin RM. (1977) Controlled and automatic human information processing I: Detection, search and attention. Psychol. Rev. 84:1-66.
- Simons DJ Chabris CF. (1999) Gorillas in our midst: Sustained inattentional blindness for dynamic events. Perception 28:1059-1074.
- Shiffrin RM Schneider W. (1977) Controlled and automatic human information processing II: Perceptual learning, automatic attending, and a general theory. Psychol. Rev. 84:127-190.
- Shipp S. (2004) The brain circuitry of attention. Trends Cog. Sci. 8:223–230.
- Sokolov EN. (1975) The neuronal mechanisms of the orienting reflex. In NE Sokolov OS Vinogradova (Eds), Neuronal mechanisms of the orienting reflex (217–238). New York: Wiley.
- Sperling G Weichselgartner E. (1995) Episodic theory of the dynamics of spatial attention. Psychol. Rev. 102:503-532.
- Spence C Lloyd D McGlone F Nicholls MER Driver J. (2000) Inhibition of return is supramodal: A demonstration between all possible pairings of vision, touch, and audition. Exp. Brain Res. 134:42–48.
- Taylor JG & Rogers M. (2002) A control model of the movement of attention. Neural Networks 15: 309-326
- Tipper SP Driver J Weaver B. (1991) Object-centred inhibition of return of visual attention. Quart. J. Exp. Psychol. A 43:289—298.
- Treisman AM & Gelade G. (1980) A feature-integration theory of attention. Cognitive psychology, 12 (1), 97-136.
- Varela F Lachaux J-P Rodriguez E Martinerie J. (2001) The brainweb: Phase synchronization and large-scale integration. Nat. Rev. Neurosci. 2:229–239.
- Ward LM. (1994) Supramodal and modality-specific mechanisms for stimulus-driven shifts of auditory and visual attention. Can. J. Exp. Psychol. 48:242—259.
- Ward LM. (2003) Synchronous neural oscillations and cognitive processes. Trends Cog. Sci. 7:553–559.
- Wickens CD. (1984) Processing resources in attention. In R Parasuramen DR Davies (Eds) Varieties of Attention (63-101). Orlando: Academic.
- Wolfe JM Cave KR Franzel SL.(1989) Guided search: An alternative to the feature integration model for visual search. J. Exp. Psychol.: Hum. Percep. Perform. 15:419–433.
- Woodman GF Luck SJ. (1999) Electrophysiological measurement of rapid shifts of attention during visual search. Nature 400:867–869.
- Wright RD Richard CM. (2003) Sensory mediation of stimulus-driven attentional capture in multiple-cue displays. Percep. Psychophys. 65:925–-938.
- Wright RD Ward LM. (2008) Orienting of Attention. New York: Oxford University Press.
- Yantis S Hillstrom AP. (1994) Stimulus-driven attentional capture: Evidence from equiluminant visual objects. J. Exp. Psychol.: Hum. Percep. Perform. 20:95–107.
- Yantis S Jonides J. (1984) Abrupt visual onsets and selective attention: Evidence from visual search. J. Exp. Psychol.: Hum. Percep. Perform. 10:601–621.
- Yantis S Serences JT (2003) Cortical mechanisms of space-based and object-based attentional control. Curr. Opin. Biol. 13:187-193.