What is attention?

Attention involves selection from all sensory modalities or from memory and is not purely visual. However, visual attention should be consistent with imaging and behavioral studies that have defined attention as involving three networks related to its functions irrespective of the modality of input (Posner & Petersen, Reference Posner and Petersen1990; Petersen & Posner Reference Petersen and Posner2012). An alerting network arises in the locus coeruleus, which is the source of the brain’s norepinephrine, and involves obtaining and maintaining a high level of vigilance to react to incoming stimuli. The orienting network connects the frontal eye fields to the superior parietal lobe and the temporal parietal junction. The orienting network connects to the ventral visual stream (object recognition) by common oscillations in the process of selecting a visual object (Womelsdorf & Fries, Reference Womelsdorf and Fries2007) and to other sensory areas which are the sites of selection by attention to auditory, touch, or other non-visual modalities. The orienting network called frontal-parietal network in imaging studies is the source of attentional effects that can influence various sensory modalities (Corbetta & Shulman, Reference Corbetta and Shulman2002). An executive network is called the cingulo-opercular network in imaging studies. The executive network is involved in more strategic functions that include the resolution of conflict between brain networks in tasks like the Stroop effect (Petersen & Posner, Reference Petersen and Posner2012). Much is unknown about how these networks operate separately or together.

Visual Orienting

The target paper discusses only orienting phenomena such as crowding and binding within the visual modality and correctly indicates that many studies are confounded between purely sensory and possible attention effects. Covert orienting to sensory stimuli involves portions of the orienting network that are the same irrespective of the sensory modality involved (Petersen & Posner, Reference Petersen and Posner2012). Thus, it is useful to distinguish between brain areas that are the source of covert shifts from those sensory sites on which orienting operates. Imaging studies show the brain areas involved include the temporal parietal junction, superior parietal lobe, and frontal eye fields which are all important in covert orienting.

Orienting of attention is most clearly involved in tasks where a cue gives priority to the location where a target may or may not occur (Posner, Reference Posner1980). Evidence that foveal location is not critical to covert orienting is that a foveal cue improves reaction time (RT) by the same amount as does a peripheral cue (Posner, Reference Posner1978). Studies demonstrate there is as much improvement in RT when the person moves attention from a cue 9 degrees to a target in the fovea as there is in the reverse movement. Orienting of attention to visual stimuli in primates may be overt involving head and eye movements or be done by a covert shift involving no observable motor output. In primates, an overt orienting movement usually places the fovea on the target location. However, rodents without a fovea show similar faster reaction times when a cue produces covert orienting to the target location (Beane & Marrocco, Reference Beane, Marrocco and Michael2004). This means that the fovea is not as critical for the priority of a stimulus as it is for acuity. However, even spatial resolution can be influenced covertly by cuing (Yeshurun & Carrasco, Reference Yeshurun and Carrasco1998).

There is controversy over whether crowding, binding, and other phenomenon are purely sensory or whether they require modulation from orienting areas of the frontal and parietal lobes. This is a purely empirical question, which is not advanced by refusing to recognize that top-down attention influences that have been shown to be critical to the priority given to a sensory stimulus (Posner & Petersen, Reference Posner and Petersen1990)

An important model task, cited in the target article (Treisman & Gelade, Reference Treisman and Gelade1980), argues that the binding together of, e.g., color and form requires attention does not seem to be correct. In one study it was shown that color and form remain statistically independent during visual processing (Isenberg, Nissen, & Marchak, Reference Isenberg, Nissen and Marchak1990) Moreover, cells that receive both color and form information exist even in primary visual cortex, but they do not lead to an integrated representation (Seymour et al., Reference Seymour, Clifford, Logothetis and Bartels2010). The integration may take place in the executive attention network or in the pathways that connect executive attention to memory (Posner et al., Reference Posner, Weible, Voelker, Rothbart and Niell2022).

To ignore what is known about attention, as suggested by the target article, would cause us to lose sight of the brain areas involved in alerting, orienting, and executive attention. Understanding attention also helps to design empirical studies that distinguish between purely sensory effects and those that involve further modulation. Moreover, we could lose our current understanding of how attention influences different sensory and memory areas and provides a foundation for understanding how top-down information influences sensation that does not occur entirely within the sensory systems.