Is the premotor theory of attention essentially about pre-reflective intentionality?

The deflationary treatment vs. the substantial theories of attention

Over the course of the past few centuries, attention has received an ambiguous treatment which has been chronicled in the philosophical and psychological literature. Following the debates on various aspects of attention, one can also discern two different fundamental stances which have been taken by philosophers and, explicitly or implicitly, by experimental psychologists on the overall cognitive value of this category (Hatfield, 1995; Mole, 2011). One stance treats attention as a phenomenon with explanatory work to do and for which a substantive independent theory needs to be given, even as “the nerve of the whole psychological system” (Titchener, 1908, p. 173). The second stance is a deflationary treatment which denies the substantial character of attention both as a causal-explanatory factor and as a phenomenon needing to be explained. A centuries-old tradition of dismissing the epistemological status of this category can be seen in Bradley’s (1886) claim, according to which the question of attention is actually artificial. “Any function whatever of the body or the mind will be active attention … There is no primary act of attention, there is no specific act of attention, there is no one kind of act of attention at all” (p. 316). A consequence of this view is that no very substantive theory of attention is needed once we have a theory of more essential cognitive processes in place. Those two approaches to attention’s explanatory remit can still be seen to resonate in modern attention labs.

One of the influential, albeit controversial (Smith & Schenk, 2012), modern accounts of attention which appears to follow the tradition of the deflationary approach to it is the Premotor Theory of Attention (PToA). It challenges the idea that attention is a distinct, i.e., modally independent, psychological entity. According to its authors, what we call visual attention is functionally equivalent to motor preparation, i.e., a program for goal-directed actions such as eye movements. That is why they predicted that it would be difficult to discover specific neural correlates of visual attention as an independent, specific cognitive phenomenon (Craighero & Rizzolatti, 2005; Rizzolatti, Riggio, & Sheliga, 1994).

Identifying the cognitive process of attention with sensori-motor machinery can be considered as embedded in the idea of embodied cognition, although, to our knowledge, the authors do not seem to explicitly subscribe to it. The vocabulary adopted by the PToA seems to bear a particular resemblance, at least on the surface, to Maurice Merleau-Ponty’s (1945/1962) notion of pre-reflective intentionality. The concept of intentionality as “reference to a content, direction towards an object” (Brentano, 1874/1973, pp. 88–89) was introduced to contemporary discussions by Brentano and was adopted by Husserl (1913/1962). According to Merleau-Ponty, however, this basic cognitive act of directing derives from bodily motility that makes it possible for the moving body-subject to reach out to things in the world and to reflect on them. The French phenomenologist prefers to define intentionality in terms of its directness towards objects and claims that there can be intentional acts which are not of or about anything specific. This is in contrast with Brentano’s or Husserl’s accounts, according to which intentional acts are of or about something and that is why they always contain an object within themselves (see Reuter, 1999).

Similarities in vocabulary between the PToA and the notion of pre-reflective intentionality raise the question concerning their possible mutual meta-theoretical relations which may sound particularly intriguing due to their different epistemological backgrounds: experimental and phenomenological. So, do they really share a common conceptual ground or, despite their apparent semantic resemblance, are they conceptually autonomous? Furthermore, given the initially limited extent of the PToA as a theory of visual attention and that the notion of pre-reflective intentionality is supposed to refer to the basic act of human cognition, this question could be expanded on to see if the former can be accommodated and brought into reflective equilibrium with the latter as a more general, and thus overarching, account of attention. In this epistemological state of affairs, the main predictions of PToA could be possibly reconstructed in terms of the phenomenological notion of pre-reflective intentionality to arrive at the embodied subject’s spatial orienting, also within one’s mental content, as an essential characteristic of the phenomenon which is generally called attention.

Thus, we develop two leading questions in this paper: (a) can the main tenets of PToA be essentially reconstructed in terms of Merleau-Ponty’s phenomenological notion of pre-reflective intentionality, which is constituted in the spatial motility of the body-subject, and since the bodily motility is meant to be at the root of all forms of intentionality, (b) can the PToA be expanded to account for all forms of attention?

In reverse order, it would also be worth seeing whether experimental findings could be utilized to address the controversy between Merleau-Ponty’s and Brentano’s (Husserl’s) understandings of intentional acts as either directed towards or being rather of or about something.

We shall address the leading questions of this article by adopting the following order. We will start by introducing the PToA as embedded in the idea of embodied cognition and by analyzing the theory in reference to one of the most influential accounts of human corporeality, namely Merleau-Ponty’s (1945/1962) phenomenology of perception for the perspective of reconstructing the former in terms of his notion of pre-reflective intentionality. Having brought the PToA into equilibrium with the notion of pre-reflective intentionality which is supposed to be at the core of all human cognitive activity, in the next section we will confront the PToA with experimental research to examine its potential as an overarching account of attentional processes. In other words, we will test the use of the PToA as a deflationary treatment of attention with the intention of arriving at the bodily spatial motility to be in the essence of all the cognitive processes which are generally labeled as “attentional.” Additionally, we will also consider the experimental discovery of the effect known as Inhibition of Return (IOR) to be an argument in favor of Merleau-Ponty’s understanding of intentionality versus Husserl’s (Brentano’s), namely that intentional acts are directed towards rather than being of or about something. It would be difficult, however, to draw on the research on the IOR as experimental evidence for the bodily character of pre-reflective intentionality.

By way of a conclusion concerning the two leading questions of this paper, we shall advocate the view that there is nothing that would prevent the PToA from being fitted into this epistemological framework provided by Merleau-Ponty. On the contrary, it may turn out to be an inspirational epistemological endeavor to see the PToA as an experimental approach par excellence, in the broader and more general perspective of his phenomenology of perception. At the same time, we will point to serious doubts as to whether the PToA can serve as an overarching account of the research results in modern attention labs. The doubts accordingly concern the use of the notion of pre-reflective intentionality in this regard.

Our main intention in this paper, however, is obviously not so much to resolve these questions as to provoke both experimental psychologists and philosophers (e.g., phenomenologists) into taking them up.

The premotor orienting of attention as pre-reflective intentionality

According to the PToA, spatial attention is not an independent cognitive function that maintains its own epistemological identity. Introspectively, it is neither a superordinate nor even a separate modality (Rizzolatti et al., 1994, p. 238). It is linked with the activity of those cortical and subcortical circuits that are involved in coding spatial representation(s) as well as programming motor actions. The authors of the theory (Rizzolatti, Riggio, Dascola, & Umiltà, 1987) formulated it on the basis of experimental data showing that responses to stimuli presented in the cued hemifield are faster than those to the uncued one, even when the distance between cue and stimulus in the cued hemifield is greater. This pattern was dubbed the meridian effect and has been explained in terms of a temporal cost of reprogramming the vector of eye movements, resulting in the latency of the saccade in the opposite direction being longer. In other words, covert attention depends primarily on the preparation of motor activity, i.e., some readiness to execute eye movements in a certain direction. Furthermore, according to the authors, “The condition in which action is ready but its execution is delayed corresponds to what is introspectively called spatial attention” (Craighero, Fadiga, Rizzolatti, & Umiltà, 1999, p. 1673). One could then say that spatial attention is not so much a consequence of a program or a plan for motor actions directed at goals located in space as it is motor preparation as such. It is functionally equivalent to motor preparation, i.e., a program for goal-directed actions such as eye movements (Rizzolatti et al., 1994).

The PToA has been developed not only in the studies of single eye movements but of sequential saccades as well (Baldauf & Deubel, 2008). It also finds support in data considered as evidence that attentional benefit for stimuli appearing in a cued location decreases considerably if a saccadic eye movement in this direction is restricted, even though the retinal stimulation is exactly the same, i.e., the stimuli are displayed fully within sight. The only difference is that the eye movement toward the to-be-attended location cannot be prepared and executed, resulting in poorer attentional performance there (Craighero, Nascimben, & Fadiga, 2004). On the basis of experimental data, spatial attention is considered to be affected not only by the programming of eye movements but also by the preparation for grasping (Craighero et al., 1999). The authors then suggest that the PToA can be extended to explain the orienting of attention to spatial locations, as well as to graspable objects, including sequences of hand-movements (Baldauf, Wolf, & Deubel, 2006).

On a neural level, spatial attention is supposed to result from the weaker activation of the same frontal-parietal circuits that determine motor behavior toward specific locations. The authors predict that it has been difficult to discover specific neural substrates of attention because such substrates do not exist (Craighero & Rizzolatti, 2005).

In psychology, the approach to attention which ascribes a bodily and motor character to it is not new and goes as far back as the 19th century. For example, in his seminal work The Emotions and the Will (1865), Alexander Bain addresses the question of what the will can do to fix the attention and provides the following account:

the retention of an idea in the mind is operated by voluntary muscles … there being a muscular element in our sensations, especially of the higher senses, touch, hearing, and sight, this element must somehow or other have a place in the after remembrance or idea; otherwise, the ideal and the actual would be much more different than we find them. The ideal circle is a restoration of those currents that would prompt the sweep of the eye round a real circle; the difference lies in the last stage, or in the stopping short of the actual movement performed by the organ … We can direct the currents necessary for keeping an imagined circle in the view, by the same kind of impetus as is required to look at a diagram in Euclid. (Bain, 1865, pp. 373–374)

Bain also anticipates, in this way, the discovery of those cortical areas that are responsible for storing and processing data concerning future motor actions.

The concept of corporeal directness which Bain comes up with to account also for attentional processes carried out within “ideation centres” of an “intimate nature”—as William James (1890, p. 411) would dub it—may further inspire the question concerning the potential for broadening the scope and directions for development of the PToA, on one hand, and giving attention a deflationary treatment, on the other. We are, however, inclined to think that one of the most influential accounts of human corporeality, namely Merleau-Ponty’s (1945/1962) phenomenology of perception, and his notion of pre-reflective intentionality in particular, might theoretically suit this purpose to a far greater extent. Provided, of course, that the PToA can be at all fitted into the framework of pre-reflective intentionality in the first place.

Intentionality is a complex notion that is analyzed on many levels and from various standpoints. The task of addressing the issue exhaustively is evidently beyond the scope of this text. According, however, to Brentano’s (1874/1973) classical formula of “the intentional (or mental) inexistence of an object” (p. 88), intentionality is an essential feature of mental phenomena as being directed towards objects, regardless of their ontological status. This formula reflects, by the way, the Latin etymology of the word referring to the idea of directedness or tension that arises from pointing towards or attending to some target. In other words, intentionality is the power of the mind to orient and encompass things, properties, and states of affairs at different times and in different modal ways, e.g., perceptual, rational, emotional.

Some of the leading discussions concerning Brentano’s expression “intentional inexistence” have focused on the meaning of the prefix in- in inexistence. Thus, they concern the question of how strict the relation between the intentional act and its object is. Is this relation so strict that intentionality and intentional acts can be described as of or about something that is internal to them? Or is this relation so loose that intentional acts should rather be described as directed towards something and are so independent that there are intentional acts which are not of or about anything specific? From this perspective, there is such a level or kind of intentionality which can be understood only as the act of directing (orienting) towards. The latter stance is taken by Merleau-Ponty. For the French phenomenologist, as Martina Reuter puts it:

Intentional acts are primarily directed toward the lived world as a whole and the phenomenon of intentionality is understandable only through its pre-reflective ability to be directed. This intentionality does not require the existence of any other objects—real or intentional—than the given world. (1999, p. 75)

To expand on the notion of pre-reflective intentionality, as far as we understand it, one can discern two aspects to it. The first is epistemological and refers to the question of what is its cognitive function. The second is ontological and refers to the question of what is its mode of being.

From the epistemological and functional perspective, pre-reflective intentionality should be understood as the fundamental force of a mind that directs it to objects, their properties or states of affairs, and subsequently allows it to reflect on them. The pre-reflectiveness of intentionality is therefore a preliminary stage, or a basic condition, of the process in which phenomena are being brought before or into an agent’s mind. Describing pre-reflective intentionality, Merleau-Ponty uses an example of love on its fundamental level as a directed act of awareness. It is “the impulse carrying me towards someone” or something before the nature of this impulse becomes “an object before my consciousness” and before I begin reflecting on its intentional object as noema in Husserl’s terminology (1945/1962, p. 381). Love of this kind, or in this aspect, as the example of pre-reflective intentionality, may be comprehended only as an act of attention. Martina Reuter (1999), who comments on Merleau-Ponty’s description of a pre-reflective intentional act, notices that it resembles Descartes’ (1649/1985) description of the passion of wonder (admiration):

In his Passions of the Soul Descartes regards wonder as the first of all the passions, it is experienced before we may make any judgements about the nature of the object, while other passions are evaluations of objects as beneficial or harmful (Descartes 1649/1985, 350). Cartesian wonder as well as the pre-reflective love described by Merleau-Ponty is a non-evaluative, but directed attention [emphasis added] towards something. (Reuter, 1999, p. 86)

A pre-reflective intentional act would then be considered to be an act of attention, i.e., an orienting of mental activities before they are carried out.

Merleau-Ponty does not, however, confine himself to describing pre-reflective intentionality only in terms of its general function which is directing or orienting of cognitive acts. He expands on the pre-reflectiveness of intentionality by giving an account of its ontological character, which is corporeal, and the manner in which it comes about. According to the French phenomenologist, the body is the basic and overall condition of a human being in the world. It constitutes the subject’s relationship with the lived world to such an extent that it can be comprehensively defined as the body-subject. In this way, the lived world, as it is for the human subject, is for a corporeal subject, not for a disembodied consciousness. Our bodily organization also provides the basis for the spatial character of our overall activity, which is carried out in the dimensions: “up, down”; “on, under”; “near to, far from”; and so on. It brings about the unity of the senses, motility, and intelligence.

Furthermore, the link between the body-subject and the lived world is so close that it can be dubbed the body-world relationship. The embodied subject’s “skill can be defined only by reference to its intended outcome, and the particular movements involved must be understood as solutions to [real] ‘problems’” (Keat, 1982, p. 5). As Merleau-Ponty himself puts it, “the task to be performed elicits the necessary movements … by a sort of remote attraction” and the body “surges towards objects to be grasped and perceives them” (1945/1962, p. 106). Thus, our bodies relate to the world in which objects present themselves as manipulanda, i.e., things to be acted upon. That is why pre-reflective intentionality involves an “I can” rather than an “I think.” “Bodily space and external space form a practical system, in which bodily space is the background against which objects may … become visible and function as goals for action” (Reuter, 1999, p. 73). This “exogenous attraction,” which elicits concrete movements, does not exhaust the repertoire of human comportment, however. There is still a “zone” to be carved out by the subject for “endogenous” abstract movements unless this ability is lost in extraordinary circumstances, for example, as a result of brain damage.

Merleau-Ponty’s (1945/1962) account of the bodily character of pre-reflective intentionality includes, however, also a dimension of corporeality that is of particular importance if this account is to be related to the PToA. The notion of bodily intentionality which Merleau-Ponty comes up with is largely based on his analysis of the pathological case of Schneider, which had been previously described by the German neuropsychologists Goldstein and Gelb (1918). Johann Schneider was a soldier injured by shrapnel which had penetrated some of his cortical areas. The resulting defects limited his body motility in somewhat artificial circumstances. For example, he could hardly point to his nose or to the position of the sensation in his leg where he had been touched with a ruler under medical examination. He sustained his ability to perform bodily movements in “normal,” everyday situations, however, such as taking a handkerchief from his pocket to blow his nose or a match out of a box to light a lamp, reaching towards his leg to scratch a mosquito bite. According to Merleau-Ponty, Schneider’s body was still capable of performing a large repertoire of actions with respect to stimulus, behavior, and muscular use. It was his ability to project his body’s movements in abstract circumstances that had been affected. He claims that Schneider’s inability to project a situation for his actions reveals a certain component of the bodily pre-reflective intentionality, which is somehow intermediary between the object-like body and the consciousness:

What he lacks is neither motility nor thought, and we are brought to the recognition of something between movement as a third person process and thought as a representation of movement—something which is an anticipation of, or arrival at, the objective and is ensured by the body itself as a motor power, a “motor project,” a “motor intentionality” in the absence of which the order remains a dead letter. (Merleau-Ponty, 1945/1962, p. 110)

This is how the embodied pre-reflectiveness of intentionality can be comprehended in terms of some potential, capacity, or skill that seeks articulation by movements taking place in the “plenum of the world.” Thus, it can also be defined as something even prior to the actual direction of motility. It is a “program” or “order” intertwined with the bodily constitution that, on the other hand, provides the means for this program to be actualized.

Can premotor orienting then be equated with pre-reflective intentionality? Since the adjective “pre-reflective” refers strictly to the noun “intentionality,” which is the process of orienting towards objects to reflect on them, or, using other vocabulary, to facilitate the processing of them, yes, it can. But is the premotor orienting a process which is embodied? In Merleau-Ponty’s account of the human body, all of the subject’s activity is bodily, which means this activity is his/her motility in space: actual, when movements are already carried out, and potential, when they are in the process of being projected or planned. To act in space, i.e., to orient, grasp, manipulate, means basically to do it in the dimensions: “up, down”; “on, under”; “near to, far from”; and so on. On the other hand, all objects that we orient towards grasp, manipulate upon, and exist in space as well. In other words, all human activity, including that which the dualists comprehend as mental, is geometrical. Since the orienting of attention, according to the PToA, is supposed to derive from mechanisms that are intrinsic to the circuits underlying both space and action (Craighero et al., 1999), it meets the criteria set by Merleau-Ponty (1945/1962) for being bodily.

The question of extending the PToA into a unified theory of attention

Bodily, i.e., spatial, directness appears to be the key quality of attention, according to the PToA and the notion of pre-reflective intentionality. Given, however, that pre-reflectiveness is at the root of all forms of intentionality, including on the symbolic level, the question arises as to whether the notion of pre-reflective intentionality can serve as a heuristic model referring to attentional processes, by and large. Or, from the opposite standpoint, can the basic tenet of PToA be extrapolated to attention in general? Can it account for all kinds of attention? It would mean that attention, to be articulated in terms of bodily directness, is generally brought about in the dimensions “up, down”; “on, under”; “near to, far from”; and so on, not only in the case of the accommodation or adjustment of the sensory organs but also as a top-down process within the mental content.

In order to examine the potential of PToA for being expanded into an overarching theory to account for the essence of those cognitive processes which are labeled as “attentional,” we will confront its main tenets and the experimental research inspired by this theory with the other that represents the current state of research in attention labs. It should, however, be underlined that the authors of PToA themselves do not explicitly aspire to provide a universal theory of attention nor even claim there is something like a global attentional effect. They had primarily confined their theory to account for spatial attention and extended it later to its specific aspect which is object-related attention (Craighero et al., 1999).

Taking up the leading question of this section, one can discern the three component questions it consists of: (a) Can attention be given a deflationary treatment and reduced to “nothing but …”, for example, the embodied subject’s spatial orienting, as its essential “flavor”?; (b) Can the PToA, which is primarily the theory of spatial attention, suit this purpose?; and (c) Does the PToA seem to be correct with regard to spatial attention, in the first place? Having already introduced the PToA, we will start to address these questions by pointing to some research that could be considered to support extending the initial scope of PToA in the context of deflationary treatment of attention and then confront this approach with the opposite tradition and the current state of experimental research in this regard. It is obvious, however, that, due to the spatial constraints of this paper, we will only be able to address them briefly.

The ambition for theorists of attention to give it a “nothing but…” reduction can be traced through the centuries. It was revealed in full swing by 19th-century authors such as Bain (1865) or Bradley (1886), following the 18th century, when attention had been treated as a phenomenon which deserved a substantive independent theory (see Hatfield, 1995). But the tradition of dismissing the explanatory status of the category of attention goes back even further, at least to John Locke (1689/1979). For him, attention is only a mode of thinking, lumped together with other modes such as “reverie,” “contemplation,” “study,” etc. As a consequence, the explanatory value of the term is slight. There is no need for any specific and substantive theory of attention once our theory of thinking is in place. In this manner, one is able to provide a comprehensive description of what is commonly called attention, using more insightful terms:

when the mind with great earnestness, and of choice, fixes its view on any idea, considers it on all sides, and will not be called off by the ordinary solicitation of other ideas, it is that we call “intention.” (II, 19 §1)

The tension between those two attitudes towards attention’s explanatory remit also seems to be present, explicitly or implicitly, in the arena of modern experimental psychology. One can see it, for example, by following the theoretical application of the moving spotlight metaphor which has been used to study attentional processing since at least the 1960s (see Wright & Ward, 2008). There is, however, some confusion in terms of what this metaphor is actually supposed to model: attention shifts or attention per se. Michael Posner (1980), who is widely credited with popularizing the spotlight metaphor with his seminal article “Orienting of Attention,” initially claims that “orienting to stimuli in visual space is a restricted sense of attention” (p. 4). One could, however, point out that having deflated attention itself to orienting in space, results in the phrase “orienting of attention” being in fact pleonastic. Posner later clarifies his stance and explicitly distinguishes orienting from attention as two functionally different components of attentional processing. If so, the following question then arises: what is there which is being oriented and that is repeatedly called attention? He calls this cognitive act detecting and specifies it to mean becoming “aware or conscious [emphasis added] of the stimulus” (p. 4). The perspective from which the orienting of attention is defined as the orienting of something that attention can be reduced to may result in the conclusion that the category of attention is dispensable and can be superseded by other terms. This is the view taken by Donald Broadbent (1958, 1971), the author of another prominent model of attention in terms of a bottle-neck of information intake. When he referred to attention, he preferred to use the term “selective perception” or “vigilance” instead. In his article “Task Combination and Selective Intake of Information,” Broadbent explicitly states that

“Attention” is a word in ordinary language, that can reasonably be used as a label for experiments in a particular area. Yet it has also been used as a theoretical concept, a mysterious asset or energy which is sometimes attached to human functions and sometimes not. This use of attention … is not very helpful, and avoiding the word in the title is a step towards clarity. (1982, p. 253)

Following the deflationary treatment at work, one can discern at least two intertwined basic functions, namely: making a vague and complex theoretical concept more clear and distinct and/by reducing it to the more simplified, possibly singular form, according to, by the way, the formula, Non sunt multiplicanda entia sine necessitate. From the reverse perspective, this essential framework, resulting from the process of deflation, should be extendable as much as it is possible to account for the phenomena which previously piqued interest. Given this characterization of the deflationary approaches to attention’s explanatory remit, the PToA can be seen as another way of carrying out a “nothing but …” reduction of this theoretical concept. This time it is supposed to be nothing but the preparation of motor activity in a certain direction. “The condition in which action is ready but its execution is delayed” (Craighero et al., 1999, p. 1673). As we pointed out, this idea is not new in psychology. It was already put forth in the 19th century by Alexander Bain (1865), for whom attention is nothing but “the stopping short of the actual movement performed by the organ” (p. 374). According to Bain, however, this accounts for attention, by and large. The question then arises concerning the potential of PToA to become an overarching account of attentional processes. It would be useful to look at some research in terms of the possible epistemological process of extending the scope of PToA in order to address this question.

The PToA, which was originally supposed to account for spatial attention, began to be developed on the basis of experiments focused on the preparation of eye movements (Rizzolatti et al., 1987). The theory has been employed, however, to account not only for attention used for enhancing perception, understood as an ability to detect or discriminate stimuli, but it also serves to explain the attention used for actions carried out in order to handle objects in the environment. In this latter case, attention is thought to derive not only from the preparation of eye movements but also from the preparation of other bodily organs or even their very constitution and shape. For example, the experiments carried out by Tipper, Lortie, and Baylis (1992) show that attention depends on the organization of arm–hand movements and is only captured by stimuli located within the arm trajectory necessary to execute a pointing response. Tucker and Ellis (1998) demonstrated that reaction times towards common graspable objects are faster with the hand (right or left) which is more suited to grasp them. Also similar is the observation that shaping the hand in a way that makes it suitable for handling an object oriented in a certain way results in faster reaction times towards that object when oriented in the same way on a screen (Craighero et al., 1999). The act of the motor preparation of attention can also take a more complex form and require “ideation centres,” as William James would call them, or, to put it plainly, a greater amount of imagination and anticipation. ¹ Adamo and Ferber (2009) demonstrated that the presentation of a tool results in attentional enhancement towards subsequently presented objects that are consistent with the actions afforded by the tools. The comportment preparation, as a sequential attentional process at work here, must thus have involved recognizing the tool, excogitating the action to be carried out with it, and its expected purpose (or the final cause in Aristotle’s terms), all in the “ideation centres” of the mind. It is, by the way, in line with our knowledge of the functional anatomy of the premotor cortex as a narrow region between the prefrontal and motor cortices. It is supposed to be responsible not only for directly commanding the initiation of a movement but also storing the information necessary for intention and the planning of motor actions or even directing the course of abstract operations.

Another example to demonstrate that the act of orienting which is bodily would be at the essence of attentional processes, also within the inner mental content, would be the phenomenon of hemineglect, unilateral neglect, hemispatial neglect, or spatial neglect. It is a disabling condition following brain damage in which patients fail to acknowledge, respond, report, or orient to stimuli and events occurring on the contralesional side of their personal or extra-personal space. The phenomenon is thought to stem from difficulties in allocating attention, a disorder of space representation, or a damage to pre-motor circuits (Deouell, Bentin, & Soroker, 2000). What seems, however, to be of particular relevance in the present context is that this neglect, apart from affecting visual, auditory, and somatosensory systems, may also affect the imagination. It is then called representational neglect. In their famous experiment, Bisiach and Luzzatti (1978) demonstrated that patients with left-sided neglect were not able to report details of the place—the Piazza del Duomo that they were familiar with—off to the left side of the viewpoint which they were asked to imagine. The fact that they failed to report these details may have resulted from the patients’ inability to orient their attention towards the memories of the scenes to the left side of the subjects in their imagination.

Having followed the epistemological extension of the PToA from the orienting of attention to spatial locations to orienting of attention to graspable objects and the experiments revealing the corporeal aspect of attentional processes, one might be inclined to consider its further extrapolation and employ the essential tenets of the PToA not only to explain visual attention but attention in general. The idea of reducing vague and complex terms to more clear and distinct ones is obviously compelling. And so is the PToA as another act of deflationary treatment of attention. The problem is, however, that such an approach seems to be rather at odds with the bulk of other research in attention labs which recognize its different modalities.

For example, an account of visual attention similar to the PToA had also been proposed before by Raymond Klein (1980) in terms of the Oculomotor Readiness Hypothesis (OMHR). Furthermore, Klein (2016) also explicitly pointed out that this kind of approach to attention seems to be palpably excited by the idea of embodied cognition. OMHR gave rise to two testable predictions about the relationship between motor preparation and visual attention: (a) oculomotor responses should be initiated more rapidly toward the location that is endogenously and covertly attended and, in a reverse order, (b) locations toward which oculomotor responses are activated (prepared) should be covertly attended. These predictions were tested and the results led Klein to reject the OMHR and to conclude that endogenously prepared eye movements and the visual attentional system are to a large extent functionally independent. Although the methodological issues made the interpretation of these results controversial, they were later confirmed with the use of improved methodology (Klein & Pontefract, 1994; see also MacLean, Klein, & Hilchey, 2015). Data which further demonstrate this conclusion can also be found, for example, in a report by Born, Mottet, and Kerzel (2014) who showed that an endogenously prepared but withheld eye movement does not result in perceptual facilitation at the prepared to-be-fixated location, in contrast to trials when it cannot be stopped. Converging evidence for a partial independence between the oculomotor system and endogenous attention also comes from studies of neuropsychological patients (Gabay, Henik, & Gradstein, 2010; Smith, Rorden, & Jackson, 2004). They demonstrate that in cases like congenital ophthalmoplegia or Duane Syndrome, which are disorders characterized by the inability to make eye movements, the patients exhibit impaired exogenous attention but no apparent deficits in endogenous orienting.

The PToA, which is believed to be deeply rooted in neurophysiological findings (e.g., Craighero et al., 1999), seems problematic not only when accounting for the corporeal character of attention, by and large. In their extensive review, Smith and Schenk (2012) also point to serious doubts concerning one of its explicit fundamental tenets that spatial attention and motor preparation use the same neural substrates which might contribute to undermining the cognitive value of PToA as such. The authors claim, however, that it may still be retained in its limited version for exogenous attention. And to this extent the PToA can be considered as an account of corporeal cognition and meta-theoretically related to the notion of embodied pre-reflective intentionality put forth by Merleau-Ponty for whom the intentional moving body is inseparably connected to the Lifeworld.

The inhibition of return as intentionality which is not about something

According to Merleau-Ponty, as we have already indicated, intentional acts are primarily directed towards the Lebenswelt as a whole. Being the Lifeworld, or the lived world, it is constantly changing. That is why it demands the constant engagement of cognitive functions to keep up with the changes that are potentially relevant for the subject. From this perspective, pre-reflective motility can be comprehended not only in terms of the potential of the body to perform a specific comportment. It can also be accounted for as an actual bodily motility, e.g., the continuous shifting of the sensory organs around in search of all possible events that might turn out to be relevant for the organism. This kind of constant motility would be considered as a readiness or alertness that expresses itself in the continuous redirecting, or reorienting, of attention. It could therefore be called pure intentionality, which is about everything or nothing in particular.

Such an idea of pure intentionality may be taken into consideration with regard to the controversy between two different understandings of intentionality: Merleau-Ponty’s and Husserl’s. The dispute concerns the noesis-noema structure. Husserl (1913/1962) follows the approach introduced by Brentano (1874/1973), according to whom intentional acts are of or about something. In other words, intentionality can be comprehended as “reference to a content” and mental phenomena defined as containing “an object intentionally within themselves” (pp. 88–89). According to Merleau-Ponty, there are intentional acts which are not of or about anything specific. For the French phenomenologist, intentionality can rather be defined by its directness which derives from a bodily spatial motility (see Reuter, 1999, p. 69). Can science be used to advocate for one of those accounts? Do we know a cognitive process that seeks articulation by intentionality as the orienting of attention towards anything specific? And, does it meet the criteria of being bodily?

We would venture to discuss whether the cognitive process which meets these criteria can be an effect called Inhibition of Return (IOR). The effect was discovered and presented by Posner and Cohen (1984). In simple terms, IOR is the inhibition of reactions to objects appearing at recently cued or attended locations. These slower and less accurate reactions are supposed to result from shifting attention away from something. More precisely, under laboratory conditions called the cuing task paradigm or the cost and benefits paradigm, the onset of the cue in the periphery leads automatically to faster and more accurate responses to targets at this location than at the opposite one. This seems to be in accordance with common sense, since it can be easily explained by the capture of attention by the cue. However, such a processing enhancement takes place only at short stimulus onset asynchronies (SOAs). At longer SOAs (approximately 300 ms) the diverse effect occurs and the efficacy of detection responses to targets at the cued location drops and increases at the uncued location. In other words, orienting visual attention to a location that has been previously attended to is, after this time, more difficult than to other ones. IOR has been used as a frame of reference to investigate attentional processes of different modalities as well as the conditions that may turn out to subserve attention (Lupiáňez, Klein, & Bartolomeo, 2006). Posner himself hoped that the study of the effect would contribute to our knowledge of visual attention as a model for how attention might operate in other domains (see Wright & Ward, 2008, p. vii). Thus, abnormal patterns of IOR are supposed to indicate attentional deficits in patients with obsessive-compulsive disorder (Rankins, Bradshaw, Moss, & Georgiou-Karistianis, 2004) which are explained in terms of difficulties in disengaging from actions. Abnormal IOR patterns have been found in cases of schizophrenia (Huey & Wexler, 1994) or Alzheimer’s disease (Langley, Fuentes, Hochhalter, Brandt, & Overmier, 2001). There are also findings that suggest that the variance of IOR may depend on normal developmental processes like the maturation of the eye-movement system in young children (Harman, Posner, Rothbart, & Thomas-Thrapp, 1994) and aging at later stages of life (Bao, Zhou, & Fu, 2004).

What is IOR, however, in its essence? Although the question generates controversial assumptions (e.g., see Hunt & Kingstone, 2003), the effect of IOR is canonically explained as an inhibitory bias against returning attention to previously attended locations (e.g., Dukewich & Klein, 2015; Lupiáňez, Martín-Arévalo, & Chica, 2013). Following the general metaphor of attention as a spotlight, it is engaged in a given location and, after having tagged it, disengages to be shifted and engaged elsewhere, and so on. IOR is therefore explained as the cost of detecting the appearance of new attention-capturing information in locations where attention has already been captured by previous events. Looking at attention through the lens of the IOR effect, it turns out to be a never-ending process of novelty seeking. If we then reformulate attention into the categories of the notion of pre-reflective intentionality, it would be understood as the habit of constant bodily reorienting, or redirecting, at anything specific, but only potential objects. This would be in line with Merleau-Ponty’s understanding of intentionality rather than Husserl’s. More support for such a perspective in which there is intentionality that is not about something comes from experiments showing that the subject’s attention can be shifted to a place before anything has been presented there (Posner, 1980). Mack and Rock (1998) take cases like these to argue that there is attention without perception. If so, and in terms of the phenomenological concept of intentionality, we are inclined to call it pure intentionality.

Yet, if IOR was to be reciprocally related to the conceptual framework provided by Merleau-Ponty, it should also reveal its bodily nature and be apt to be explained in terms of the PToA. This is then the question of the ontological status of the process which is at work here. Michalczyk, Paszulewicz, Bielas, and Wolski (2018), using an eye abduction technique, showed that restricting eye movement results in IOR attenuation. Since the eye movement restriction affects the motor programming which is responsible for the movement’s preparation, the results are supposed to be in line with the PToA. The problem is, however, that such results are again at odds with most research on the mechanism(s) underlying IOR.

There seems to be a rather consistent general consensus that one can distinguish two “flavors” or types of IOR (Dukewich & Klein, 2015; Hilchey, Klein, & Ivanoff, 2012): motoric (or motor/oculomotor/output) which engages the oculomotor system and attentional/perceptual (or cognitive/attentional or sensory/perceptual/input) occurring when the oculomotor system is quiescent. One may then conclude that only one of them can be characterized as being bodily. For instance, in the research that we have already cited here (Smith et al., 2004) the authors had demonstrated and later confirmed (Smith, Jackson, & Rorden, 2009) that the patients’ inability to make eye movements, as it is in the case of congenital ophthalmoplegia, does not affect IOR, indicating that the effect can be generated independently of the oculomotor system.

As experimental evidence considered in connection with Merleau-Ponty’s understanding of intentionality vs. Brentano’s and Husserl’s, these results again suggest a split conclusion. From a functional perspective, the effect of IOR, regardless of its “flavor,” supports the notion that there are intentional acts which are not about or of anything in particular but are rather directed towards somewhere. From an ontological perspective, however, it appears that these acts do not have to necessarily be bodily.

Conclusion

William James’ famous 1890 statement that “Everyone knows what attention is” (p. 403) sounds controversial in the face of the two competing perspectives that have been used to approach this issue for centuries. One of them treats attention as an independent cognitive function that maintains its own epistemological identity. The other is a deflationary treatment that dismisses the explanatory status of the category of attention and claims that it can (and should) be reduced to and superseded by more substantial terms. In this theoretical and meta-theoretical context, we regarded the Premotor Theory of Attention, which is one of the most prominent, albeit controversial, modern accounts of attention, as an exemplar of a deflationary approach to it. According to the PToA, what we call spatial attention is equivalent to motor preparation, i.e., a program for goal-directed actions such as eye movements and reaches. This conceptualization occurred to us as embedded in the idea of embodied cognition and bearing a particular resemblance, at least on the surface, to Merleu-Ponty’s notion of pre-reflective intentionality. The French phenomenologist defines it in terms of the subject’s directness towards events in the world which derives from bodily motility and is at the root of all forms of intentional acts. Apparent similarities between both of those accounts, particularly intriguing due to their different epistemological backgrounds, i.e., experimental and phenomenological, inspired us to come up with the question of whether the main tenets of PToA can essentially be reconstructed in terms of the notion of pre-reflective intentionality. And, given the officially limited extent of the PToA as a theory of spatial attention and the fundamental character of pre-reflective intentionality, we expanded the question to also examine the potential of PToA as a possible overarching account of attentional processes by confronting this idea with the current state in attention labs.

In reverse order, we also sought to utilize experimental findings in order to address the controversy between Merleau-Ponty’s and Brentano’s (Husserl’s) understandings of intentional acts as either directed towards or being rather of or about something.

In regard to both of these leading questions, we have come up with a split conclusion. Although we are inclined to advocate the view that there is nothing that would prevent the PToA from being fitted into the epistemological framework provided by Merleau-Ponty, we point to serious doubts concerning the potential expandability of PToA into a universal theory of attention. Taking experimental research on attentional processes into consideration, and the effect of IOR in particular, we also conclude in favor of Merleau-Ponty’s (1945/1962) understanding of intentionality versus that of Brentano and Husserl, namely that intentional acts are directed towards rather than being of or about something. Here again, however, it would be difficult to use these research findings in order to defend the necessarily bodily character of pre-reflective intentionality.