The term “attention” has a broad set of meanings in everyday language, but they are all related in some way to the concept of focusing mental processes. Focusing on one task to the exclusion of others is often called “concentration” in everyday language. This kind of attention is typically studied under the heading of executive control. A good example is the Stroop task, in which the challenge is to perform a relatively unpracticed task (saying the name of a color) while suppressing interference from a highly practiced task (reading a word). This variety of attention is discussed in part IX, on higher cognitive functions (chapters 69–76).

Focusing on one source of sensory inputs to the exclusion of others is termed selective attention or, as a shorthand, simply “selection.” A good example is the visual search task, in which the challenge is to find a target object while suppressing interference from distractor objects that may be perceptually similar to the target. This is the variety of attention that is the focus of this section. Although selective attention plays a key role in many sensory modalities, the largest body of work has been in the visual system because our detailed knowledge of the anatomy and physiology of this sensory system provides solid footing for the study of the more ephemeral topic of attention. Visual attention is therefore the main focus of the chapters in this section.

Perhaps the most fundamental distinction in the study of attention is between the control of attention and the implementation of attention. Attentional control processes are responsible for taking general task instructions (e.g., find a grapefruit on the left side of this photograph) and converting these into a bias toward an appropriate set of features (e.g., colors, shapes, locations, etc.). Once a set of potentially relevant features, objects, or locations has been found, attentional implementation processes are responsible for ensuring that these features, objects, or locations receive preferential processing. In the context of the classic spotlight metaphor, control processes are responsible for directing the attentional beam, and attention is implemented by means of the illumination of objects by the beam.

As discussed in Treisman's chapter on the basic behavioral phenomena of attention, early research on attention focused on the implementation of attention (chapter 12). This research initially asked whether attention operates at an early, perceptual stage or a late, postperceptual stage, but cognitive neuroscience research has shown that these are not mutually exclusive alternatives. Instead, attention can operate at different stages depending on the nature of the stimuli and task. Moreover, recent research has moved beyond coarse distinctions between perceptual and postperceptual stages and is now examining how attention operates in different ways within the dozens of different areas of the visual processing pathway. This new research is reviewed by Kastner, McMains, and Beck in chapter 13, on neuroimaging research, and by Maunsell in chapter 19, on single-unit recordings.

Research on the implementation of attention has also become more precise in identifying the nature of attentional modulations of visual sensory responses. Two major principles have now been supported by many forms of converging evidence. First, attentional selection depends strongly on the degree of competition. As described in the chapters by Kastner, McMains, and Beck and by Maunsell, attention has its strongest effects when task-irrelevant information competes with the processing of task-relevant information. For example, attention effects are often observed to increase as information moves into higher stages of the visual processing pathway, where receptive fields are larger and are therefore likely to contain task-irrelevant objects as well as task-relevant objects. A second major principle is that attention often operates as a gain control, increasing the effective contrast for attended stimuli without changing sensory tuning curves. This point has been made most clearly in the psychophysical and single-unit studies reviewed by Maunsell and in the event-related potential (ERP) studies reviewed in chapter 15 by Hopf, Heinze, Schoenfeld, and Hillyard. We are also now beginning to understand how these effects may arise from the microcircuitry and temporal dynamics of visual cortex, as reviewed in chapter 20 by Womelsdorf and Fries on the role of neural synchronization in attention.

The control of attention is also described in considerable detail in this section. Much of this work takes place within the concept of a network of frontal, parietal, and superior temporal areas that control the operation of attention. The chapters in this section make the case for specializations in attentional control within these interconnected brain regions. In chapter 17, Karnath describes the detailed anatomy of a network of areas surrounding the Sylvian fissure involved in the control of attention, noting an interesting correspondence between right-hemisphere areas that are involved in attentional control and left-hemisphere areas that are involved in language and imitation. Chapter 14 by Corbetta, Sylvester, and Shulman complements this, describing how attentional control networks receive information from subcortical systems involved in affect and motivation and how different dorsal and ventral frontal-parietal networks are involved in the control of sensory and motor systems. Mangun, Saron, and Walsh describe how frontal and parietal attentional control networks interact with frontal cortical systems, such as the anterior cingulate cortex, that are involved in conflict detection, error monitoring, and online behavioral adjustments (chapter 16).

Research on the control of attention has also focused on the elementary “units” of selection. That is, does attention select spatial locations, nonspatial features, or whole objects? As reviewed in Treisman's chapter, this topic has been the source of much dispute in the cognitive literature for more than two decades, but cognitive neuroscience research has shown that these are not mutually exclusive alternatives. Attention can be allocated to the features of objects, to the locations occupied by objects, and directly to the objects themselves. This finding can be seen in neuroimaging studies (as described by Beck and Kastner and by Corbetta et al.), in single-unit studies (as described by Maunsell), and in ERP studies (as described by Hopf et al.). Perhaps the most impressive evidence, however, comes from the lesion patients described by Robertson in chapter 18. When spatial processing is severely disrupted in these patients, it is still possible to see clear evidence of intact feature-based and object-based attention.

Now that we have reviewed the major themes discussed by the chapters in this section, we would like to highlight three areas in which substantial progress has been made since the last edition of this book. First, although we have known for decades that a broad set of frontal, parietal, and subcortical areas play a role in the control of attention, the chapters in this section provide a much more detailed description of how these areas work both independently and in concert to control attention. New research is also looking at the microcircuitry of attention, revealing how different subclasses of neurons within an area are modulated by the operation of attention.

Second, the new research described in this section is beginning to reveal the details of how frontal and parietal areas can control the operation of attention within sensory areas. Top-down control signals have now been measured in a broad variety of attention tasks, and stimulation techniques have been used to show that activity within control areas can directly modulate sensory responses within visual cortex. A major unknown, however, is how these signals lead to the changes in neural synchrony that now appear to play an important role in the implementation of selection.

Finally, the chapters in this section describe new insights into the allocation of attention to nonspatial features, such as color and direction of motion. Recent studies have shown that attention can be directed to specific feature values across the visual field, and not just at attended locations. Indeed, the operation of feature-based attention sometimes precedes and guides the allocation of space-based attention. We expect that the next five years will lead to new insights into how these different varieties of attention work together in the service of perception and behavior.