Effects of Awareness on the Control of Attention Taylor W. Webb, Hope H. Kean, and Michael S. A. Graziano Abstract ■ Previous studies show that it is possible to attend to a stim- suppression of attention to a task-irrelevant stimulus, was re- ulus without awareness of it. Whether attention and awareness duced. Third, attention was more driven by the luminance con- are independent or have a specific relationship, however, re- trast of the stimulus. These findingsaddtothegrowing mains debated. Here, we tested three aspects of visual attention information on the behavior of attention with and without aware- with and without awareness of the visual stimulus. Metacontrast ness. The findings are also consistent with our recently proposed masking rendered participants either subjectively aware or not account of the relationship between attention and awareness. aware of the stimulus. Attention drawn to the stimulus was mea- In the attention schema theory, awareness is the internal model sured by using the stimulus as a cue in a spatial attention task. of attention. Just as the brain contains a body schema that We found that attention was drawn to the stimulus regardless of models the body and helps control the body, so it contains an whetherornotpeoplewereaware of it. However, attention attention schema that helps control attention. In that theory, in changed significantly in the absence of awareness in at least the absence of awareness, the control of attention should suffer three ways. First, attention to a task-relevant stimulus was less in basic ways predictable from dynamical systems theory. The stable over time. Second, inhibition of return, the automatic present results confirm some of those predictions. ■ INTRODUCTION nal model occurs, movement of the body is still possible It is now well established that attention and awareness but suffers characteristic deficits in control. If the internal are separable. People can attend to a stimulus in the ab- model is missing, then the arm is less stably maintained in sence of awareness of that stimulus (Norman, Heywood, a task-relevant state, is less able to transition away from a & Kentridge, 2013; Hsieh, Colas, & Kanwisher, 2011; no-longer-desired state, and is more easily perturbed by Kentridge, Nijboer, & Heywood, 2008; Koch & Tsuchiya, external forces (Scheidt, Conditt, Secco, & Mussa-Ivaldi, 2007; Jiang, Costello, Fang, Huang, & He, 2006; Tsushima, 2005; Graziano & Botvinick, 2002; Wolpert, Ghahramani, Sasaki, & Watanabe, 2006; Lamme, 2004; Lambert, Naikar, & Jordan, 1995). In the attention schema theory, the rela- McLachlan, & Aitken, 1999; McCormick, 1997; Lambert, tionship between attention and awareness is similar to the Beard, & Thompson, 1988). Yet attention and awareness relationship between the body and the body schema. In are not entirely independent, given that interactions be- this theory, without awareness, attention should still be tween them have been reported (Tsushima et al., 2006; possible, but the control of attention should suffer. It Lambert et al., 1999; McCormick, 1997). The specific rela- should be less stably maintained in a task-relevant state, tionship between attention and awareness, however, has less able to transition away from a task-irrelevant state, remained uncertain. and more easily perturbed by external influences. Recently, we proposed the attention schema theory, a The purpose of this study was twofold. First, because possible account of the relationship between attention relatively few studies have directly compared attention in and awareness (Webb & Graziano, 2015; Graziano, the presence and absence of awareness, we sought to col- 2013, 2014; Graziano & Webb, 2014; Kelly, Webb, Meier, lect basic information on how attention to a visual stimu- Arcaro, & Graziano, 2014; Graziano & Kastner, 2011). In lus behaves when participants are aware of the stimulus the theory, awareness is the internal model of attention and when they are not aware of it. The data may be useful or the attention schema. A basic principle of control theory in informing any hypothesis about the relationship be- is that a control system benefits from an internal model of tween attention and awareness. the thing to be controlled (Camacho & Bordons Alba, A second, more specific purpose was to test predic- 2004). For example, the brain constructs a body schema, tions of the attention schema theory. Three specific pre- an approximate internal model of the body, to help con- dictions were tested. Each prediction was made by trol movement. When misalignment or error in the inter- applying the basic concept of an internal model in dynam- ical systems control to the case of attention. The first pre- diction concerns stability. In control theory, the internal Princeton University model helps to stabilize the system in a desired state © 2016 Massachusetts Institute of Technology Journal of Cognitive Neuroscience 28:6, pp. 842–851 doi:10.1162/jocn_a_00931 (Camacho & Bordons Alba, 2004). The loss of an internal metacontrast masking was timed to prevent awareness model leads to a reduction of stability. In the case of the of the cue, IOR should be reduced or absent. This exper- arm, without an internal model, the system is less able to iment is of interest beyond testing a specific prediction of monitor and thus maintain a stable desired arm position. the attention schema theory. Whether IOR occurs in the The arm wobbles. If the attention schema theory is cor- absence of awareness is a basic question about the behav- rect, then without awareness of the stimulus, attention ior of attention, and it has been studied and debated drawn to a task-relevant stimulus should become less sta- before (Ivanoff & Klein, 2003; Lambert et al., 1999; ble in time. Attention should wobble. To test this pre- McCormick, 1997). Previous experiments arguably did diction, in Experiment 1, we directly compared attention not use directly comparable stimuli on interleaved aware to a visual stimulus with and without awareness of the and unaware conditions to determine quantitatively how stimulus. We used a Posner spatial attention paradigm IOR may change. We hope therefore that this study will (Posner, 1980) in human participants to measure atten- add to that literature. tion drawn by a brief visual cue. Attention was measured The third prediction concerns perturbations by exter- at 5 time points in the first 600 msec after cue presenta- nal influences. In control theory, without an internal model, tion to provide a temporal profile. Metacontrast masking the item being controlled is less internally stabilized and (Breitmeyer & Öğmen, 2006) was applied such that par- more affected by external drivers. In the case of the ticipants reported being subjectively aware of the cue on arm, without an internal model of the arm to help stabilize some trials and unaware of it on other, interleaved trials. it, the control mechanism is less able to resist external In this manner, we tested how the time course of atten- forces. The arm becomes more easily perturbed. In the tion changed depending on the presence or absence of case of attention, the external influences are bottom–up, awareness of the cue. The specific prediction of the atten- sensory-driven factors such as the luminance contrast of tion schema theory was that attention drawn by the cue a visual stimulus. If the attention schema theory is cor- would show greater stability through time in the presence rect, these external drivers of attention should have a of awareness and show significantly greater temporal in- greater impact in the absence of awareness than in the stability in the absence of awareness. presence of awareness. Attention normally depends on The second prediction concerns the control of move- stimulus contrast, but that dependence should have a ment away from a task-irrelevant or nondesired state. In steeper slope when awareness is absent than when aware- control theory, the loss of an internal model compro- ness is present. To test this prediction, in Experiment 3, mises a control system’s ability to register when it is in we again used a Posner paradigm to measure attention a nondesired state and therefore impairs the system’s drawn to a visual cue. To focus on bottom–up attention, ability to actively leave or avoid that nondesired state. we tested attention 50 msec after the cue onset. The In the case of the arm, without an internal model, the luminance contrast of the cue was varied, and the effect system cannot register that the arm has entered a task- of contrast on attention was measured. The specific pre- irrelevant configuration, and therefore, compensatory diction of the theory was that stimulus contrast would movements away from that nondesired state are impaired. have a greater impact on attention in the absence of In the case of attention, there is a well-studied phe- awareness of the cue than in the presence of awareness nomenoninwhichattentionactivelytransitionsaway of the cue. Once again, this experiment may have inter- from a task-irrelevant state. If attention is drawn to the est beyond testing a specific prediction of the attention onset of a task-irrelevant stimulus, that attention is typically schema theory. Many previous studies have demonstrated rapidly attenuated and can even become “negative” in the stimulus-driven attention in the absence of awareness of sense that attention briefly avoids that location in prefer- the stimulus (e.g., Tsushima et al., 2006; Lambert et al., ence for other locations, a phenomenon known as inhibi- 1999; McCormick, 1997). To our knowledge, however, tion of return (IOR; Posner, Rafal, Choate, & Vaughan, no prior experiment has tested how awareness might 1985; Posner & Cohen, 1984). If the attention schema affect the specific relationship between stimulus contrast theory is correct, then without awareness of the visual and stimulus-driven attention. The experiment can there- stimulus, IOR should be reduced or eliminated. To test fore add to the growing knowledge of how attention be- this prediction, we used the same paradigm as for Experi- haves with and without awareness.