Consciousness and Cognition 24 (2014) 12–21

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

Review

Brain signals do not demonstrate unconscious decision making: An interpretation based on graded conscious awareness Jeff Miller a,⇑, Wolf Schwarz b,* a b

Department of Psychology, University of Otago, Dunedin, New Zealand University of Potsdam, Department of Psychology, P.O. Box 60 15 53, D-14415 Potsdam-Golm, Germany

a r t i c l e

i n f o

Article history: Received 9 February 2013

Keywords: Neuroscience Consciousness Decision making Libet

a b s t r a c t Neuroscientific studies have shown that brain activity correlated with a decision to move can be observed before a person reports being consciously aware of having made that decision (e.g., Libet, Gleason, Wright, & Pearl, 1983; Soon, Brass, Heinze, & Haynes, 2008). Given that a later event (i.e., conscious awareness) cannot cause an earlier one (i.e., decisionrelated brain activity), such results have been interpreted as evidence that decisions are made unconsciously (e.g., Libet, 1985). We argue that this interpretation depends upon an all-or-none view of consciousness, and we offer an alternative interpretation of the early decision-related brain activity based on models in which conscious awareness of the decision to move develops gradually up to the level of a reporting criterion. Under this interpretation, the early brain activity reflects sub-criterion levels of awareness rather than complete absence of awareness and thus does not suggest that decisions are made unconsciously. Ó 2013 Elsevier Inc. All rights reserved.

Contents 0. 1. 2. 3. 4.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Neurobiological evidence for an epiphenomenal view of consciousness . . . . . Graded processes underlying decision making. . . . . . . . . . . . . . . . . . . . . . . . . . A reinterpretation of brain activity preceding reports of conscious decisions . General discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

12 13 15 17 17 19

0. Introduction One of the oldest unsolved problems in the philosophy of science concerns the relationship between the mind and the brain (e.g., Baars & Gage, 2010; Dehaene, 2001; Gregory, 2004; Metzinger, 2000; Milner & Rugg, 1992; Rees, Kreiman, & Koch, 2002). Naturally, new insights into this relationship would be expected from studies exploiting the massive recent technological advances in the neurosciences. What may be surprising to many, however, is the conclusion commonly drawn from these studies—namely, that decisions to initiate a movement are actually made unconsciously and merely ‘‘bubble up’’ subsequently into consciousness. As is described in detail in the next section, the evidence supporting this view is that brain ⇑ Corresponding authors. Fax: +64 3 479 8335. E-mail addresses: [email protected] (J. Miller), [email protected] (W. Schwarz). 1053-8100/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.concog.2013.12.004

J. Miller, W. Schwarz / Consciousness and Cognition 24 (2014) 12–21

13

activity associated with decision making begins before a person reports being consciously aware of having made that decision. One interpretation of this evidence is that our subjective impression of initiating decisions through an extended process of conscious deliberation is mostly an illusion—the brain has already done it before we are even aware of it—and the conscious part of intention formation is largely epiphenomenal (e.g., Libet, Gleason, Wright, & Pearl, 1983; Soon, Brass, Heinze, & Haynes, 2008; for reviews see Libet 1985, Pockett 2004, Roskies 2006, 2010 and Wegner 2002). At most, consciousness is thought to be involved in vetoing these subconsciously-generated decisions (Libet, 1985); that is, although the decision itself evolved unconsciously, its overt execution may later be consciously inhibited (but see Filevich, Kühn, & Haggard, 2013 for evidence that veto decisions are also made unconsciously). This idea of unconscious decision making is regarded by many as one of the most important contributions of recent neuroscientific work, not least because it seems incompatible with free will and thus to have far-reaching political and social implications (e.g., Lavazza & De Caro, 2010; Roskies, 2010). The thesis that decisions to move are made unconsciously has been quite controversial among both neuroscientists and philosophers (e.g., Klemm, 2010; Lavazza & De Caro, 2010; Radder & Meynen, 2013). Critics within the neuroscience community have tended to highlight specific methodological weaknesses of the key studies (e.g., Breitmeyer, 1985; Haggard & Eimer, 1999; Joordens, Spalek, Razmy, & Van Duijn, 2004; Miller, Shepherdson, & Trevena, 2011; Näätänen, 1985; Ringo, 1985; Trevena & Miller, 2002, 2013; Van de Grind, 2002; and many of the commentaries following Libet, 1985). Philosophers have generally questioned implicit assumptions about the nature of the underlying decision-making processes and claims that the reported biological measures must index them (e.g., Mele, 2009; Radder & Meynen, 2013; Smith, 2011). In this article we develop a further critique of the idea that neurobiological findings support unconscious decision making. Specifically, we focus on the claim that people are not conscious of their decision-making activity when the associated brain activity begins. We argue that this claim is based on a simplistic, all-or-none view of consciousness that is quite implausible in the light of considerable evidence, as is reviewed below, that consciousness develops in a graded manner instead. We conclude that existing neurobiological findings do not provide strong support for the idea of unconscious decision making and that they cannot do so without much more nuanced measures of conscious awareness. In Section 1 we briefly summarize two example neurobiological mind-brain studies that have been interpreted as support for the conclusion of unconscious decision making. Although these studies are superficially rather different, they provide fundamentally the same type of evidence for that conclusion. Moreover, the inferences of both studies rest in the same way on a strict all-or-none notion of conscious awareness (Fahle, Stemmler, & Spang, 2011). In Section 2, we review current theories of decision making, all of which are based on the idea of a graded evidence-accumulation process. We summarize diverse sources of empirical evidence and theoretical precedents supporting this graded view and suggest that—in light of them—it is quite plausible that people’s decisions about their own conscious states—like all other decisions—also involve a graded accumulation process. Finally, in Section 3 we revisit the results of the neurobiological studies used to argue for unconscious decision making and show that they can be interpreted quite differently if conscious awareness is conceptualized as graded rather than all-ornone. Our overall conclusion, then, is that current neurobiological findings do not unequivocally demonstrate unconscious decision making but instead may be quite compatible with people’s subjective impressions that they reach decisions through a process of conscious deliberation. 1. Neurobiological evidence for an epiphenomenal view of consciousness In a seminal study of the relation between brain activity and conscious decision making, Libet et al. (1983) asked their participants to execute a free voluntary motor act—a brisk, abrupt flexion of the right wrist or fingers. The participants were instructed to let the urge to act appear on its own at any time, without preplanning. They were also told to monitor a revolving clock, and to note and later report the time W—the earliest appearance of a conscious awareness of the specific urge, intention, decision or ‘‘wanting’’ to move. Libet et al. (1983) recorded the readiness potential (RP) as well as the electromyogram (EMG) at the activated forearm muscle. The basic finding, idealized in Fig. 1A, was that the onset of the RP preceded the reported time W by hundreds of milliseconds. Libet et al. (1983) concluded that ‘‘the brain evidently decides to initiate [. . .] the act at a time before there is any reportable subjective awareness that such a decision has taken place’’, and that the ‘‘cerebral initiation even of a spontaneous voluntary act [. . .] can and usually does begin unconsciously’’ (p. 640), and others have similarly interpreted Libet et al.’s findings as evidence that ‘‘no role appears for conscious processes in the control of action’’ (Obhi & Haggard, 2004, p. 360). Later, Libet (1985) acknowledged that conscious processes ‘‘could still have a role either in completing the initiating process (‘conscious trigger’) or in blocking its progression (‘veto’)’’ (p. 536), but he never wavered from his main message that ‘‘the initiation of a voluntary act occurs unconsciously, before a subject is aware of the wish or urge to act’’ (Libet, 2003, p. 327, italics in original). In a recent and conceptually very similar study addressing the same fundamental issue, Soon et al. (2008) asked participants to freely decide, when they felt the urge to do so, between pushing a button with the left or right index finger; at the same time, their brain activity was measured using functional magnetic resonance imaging (fMRI). A sequence of letters was visually presented with a stimulus onset asynchrony of 500 ms, and participants had to remember and later report the letter shown at the moment when their motor decision was consciously made. The authors identified frontopolar and parietal brain regions whose measured activation contained significant predictive information about the identity and timing of the response several seconds before the onsets of the letters that participants reported as having been present when they

14

J. Miller, W. Schwarz / Consciousness and Cognition 24 (2014) 12–21

Physiological Signal

A

Awareness

B

C Awareness

"Criterion"

-1000 -800 -600 -400 -200

0

Time (ms)

W

M

Fig. 1. Idealized illustrations of the possible time courses of physiological signals and conscious awareness of the urge to move in spontaneous movement tasks like that of Libet et al. (1983). (A) Physiological signals (e.g., readiness potential) preceding the times of the actual movement (‘‘M’’) and the participant’s reported time of consciously deciding to move (‘‘W’’ or ‘‘will time’’). These idealized signals illustrate the common finding that physiological signs of the movement can often be observed before the reported time of the decision to move (e.g., Fried et al., 2011, Figs. 2–6; Libet et al., 1983, Fig. 2). (B) Time course of awareness of the urge to move, according to the view that this awareness develops in an all-or-none manner. The participant reports as W the time at which the all-or-none transition occurs. (C) Time course of the awareness of the urge to move, according to the view that this awareness develops in a graded manner. The participant reports as W the time at which the awareness reaches a criterion level.

made their decisions. The authors stressed that these lead times are too long to be explained by mcere timing inaccuracies in reporting the onset of awareness, which was a major methodological criticism of Libet et al. (1983) and similar earlier studies, and they concluded that ‘‘a network of high-level control areas can begin to shape an upcoming decision long before it enters awareness’’ (Soon et al., 2008, p. 545). The study of Soon et al. (2008) exemplifies the massive recent progress in brain imaging technology and computational power, but its basic design and inferential logic are essentially identical to those of Libet et al. (1983): a reported time of consciously making a decision, W, is compared to a simultaneously obtained physiological time series. The latter is found to contain predictive information about the timing or identity of the chosen response even before W, and this early brain activity is interpreted as evidence of neural ‘‘areas that begin to prepare an upcoming decision long before it enters awareness’’ (Soon et al., 2008, p. 543). Many essentially similar studies following this basic design logic and finding comparable results have now been reported, and for convenience we will refer to these collectively as ‘‘Libet-type’’ studies. Some have even used single cell recordings as their physiological measure (e.g., Fried, Mukamel, & Kreiman, 2011). Overall, the results of many Libet-type studies like those of Libet et al. (1983) and Soon et al. (2008) indicate that brain activity correlated with a decision can be observed before a person reports being consciously aware of having made that decision. Given that a later event (i.e., conscious awareness) cannot cause an earlier one (i.e., decision-related brain activity), such results appear to indicate that consciousness does not cause the decisions to be made. Instead, one interpretation of these findings is that both the decision itself and the conscious awareness of it emerge as consequences of the prior brain

J. Miller, W. Schwarz / Consciousness and Cognition 24 (2014) 12–21

15

activity, leading to the view that decisions are made unconsciously and consciousness itself is merely an epiphenomenon (e.g., Libet, 1985). For the present purposes, the most important feature of this argument for unconscious decision making is that it treats the development of conscious awareness as an all-or-none transition at the moment, W, indicated by the participant’s time judgment (Fahle et al., 2011). As has been discussed extensively, the time of a conscious experience is ultimately a subjective judgment—one that can be made directly only by the person having the experience—and there could well be many sources of inaccuracy and bias in the measurement of W (e.g., Haggard & Eimer, 1999; Joordens, Van Duijn, & Spalek, 2002; Miller, Vieweg, Kruize, & McLea, 2010; Pockett & Miller, 2007), especially when the experimental procedure requires division of attention between internal states and external stimuli (e.g., Corallo, Sackur, Dehaene, & Sigman, 2008; Haggard, 1999). Indeed, despite rapid advances in the measurement of brain processes, there have been no important advances within the last 100 years in measuring the time of a conscious experience. We fully agree with many of the points raised concerning possible biases and unreliability in measuring W. However, we emphasize that these criticisms neglect an even more fundamental problem involving W—one that would be present even if W could be measured with perfect accuracy. Specifically, the argument for unconscious decision making implicitly assumes that W reflects an all-or-none transition from a state of no awareness to one of full awareness (Haggard, 2005), as is depicted in Fig. 1B. In contrast, consciousness may instead be graded (e.g., Seth, Dienes, Cleeremans, Overgaard, & Pessoa, 2008), with a gradual transition from no conscious awareness to full awareness, as is depicted in Fig. 1C. Some basic facts about decision making that are quite consistent with graded accounts will be reviewed in the next section, and a subsequent section will address the implications of graded accounts for possible alternative interpretations of Libet-type studies. 2. Graded processes underlying decision making There is broad agreement that some unconscious neural processes underlie and support perception, cognition, and action (e.g., Dehaene & Changeux, 2011; Grill-Spector, Kushnir, Hendler, & Malach, 2000; Van Gaal & Lamme, 2012). These processes would include, for example, early sensory (e.g., retinal) and late motor (e.g., spinal reflex) activity that could be elicited even in an unconscious individual. Moreover, even the conscious aspects of perception, cognition, and action are generally thought to go hand in hand with underlying neural activity. For example, concerning the relation of the neural and conscious processes involved in volition, Haggard (2008, p. 941) wrote ‘‘both conscious intention and action are driven by a common cause, namely the neural preparation for action.’’ Thus, the remarkable aspect of the unconscious decision making view advanced by Libet et al. (1983) and others is not that unconscious processes are involved in spontaneous movements—this point was not in doubt—but rather that the decision to move springs fully formed into consciousness after an entirely unconscious buildup. Our view, in contrast, is essentially that conscious awareness of the decision to move builds gradually and that its apparent all-or-none character in the experiments of Libet et al. (1983) and others stems primarily from the experimenter’s artificial requirement that participants identify a discrete time point dividing conscious from unconscious states.1 To support our view, in this section we review evidence that decisions are made gradually and that people have conscious access to the intermediate states of the decision processes, notwithstanding the fact that these states are driven by unconscious neural processes. Based on a wide variety of evidence, current models of both sensory and motor-related decision making have converged on the idea of a gradual evidence-accumulation process like that depicted in Fig. 1C (e.g., Bogacz, 2007; Brown & Heathcote, 2008; Gold & Shadlen, 2007; Luce, 1986; Macmillan & Creelman, 2005; Purcell et al., 2010; Ratcliff, 2006; Smith, 2000; Smith & Ratcliff, 2009; Usher & McClelland, 2001). In these models, a decision process starts from some—possibly biased—initial position, and arriving evidence is accumulated until the total evidence reaches a preset criterion, at which point the decision is reached. Gradual evidence-accumulation models of this sort are supported by both neurophysiological and behavioral findings. For example, neurophysiological studies of both single-neuron activity (e.g., Gold & Shadlen, 2000; Hanes & Schall, 1996; Ratcliff, Cherian, & Segraves, 2003) and event-related brain potentials (e.g., Gratton, Coles, Sirevaag, Eriksen, & Donchin, 1988; Leuthold, Sommer, & Ulrich, 2004) reveal a gradual buildup of neural activity prior to perceptions, decisions, and motor responses. Indeed, the gradual neural buildups seen in Libet-type studies (e.g., Libet et al., 1983; Soon et al., 2008) simply provide additional examples within the much wider literature indicating gradual buildup of neurophysiological activity. In many experimental situations, this buildup is interpreted as the neural reflection of evidence accumulation within perceptual, cognitive, and motor levels (for reviews, see e.g., Gold & Shadlen, 2007; Schall, 2001, 2003). In Libet-type spontaneous movement paradigms where there is no stimulus evidence to be accumulated, many have suggested that the gradual buildup reflects increasing neural preparation for movement (e.g., Libet et al., 1983), although Schurger, Sitt, and Dehaene (2012) argued that it could instead reflect random-walk type fluctuations in motor activation below a threshold for motor responding. Whatever neural processes they reflect, graded neurophysiological changes are perfectly consistent with the idea that awareness also develops in a graded manner. They are not sufficient to demonstrate graded changes in awareness, 1 Like e.g., Libet (1985, 2003), we limit our discussion of consciousness to the primary contents of awareness often associated with focal attention (Mangan, 1993). Although there has been extensive philosophical discussion of the elements and structure of consciousness (e.g., Gurwitsch, 1964; Husserl, 2001; Mangan, 1993), it seems clear that participants in Libet-type paradigms interpret the instructions as telling them to report the moment at which they are focally aware of deciding to move.

16

J. Miller, W. Schwarz / Consciousness and Cognition 24 (2014) 12–21

however, precisely because the relation between neural activity and awareness remains speculative (e.g., Dehaene & Naccache, 2001). Thus, it is necessary to complement the evidence of graded neurophysiological changes with corresponding behavioral studies to which we turn next. Behavioral studies address conscious awareness by focusing on people’s choices, decisions, and reports, which obviously reflect their conscious states at least to some degree (e.g., Searle, 2000). For example, evidence from chronometric studies of perceptual-motor performance strongly supports graded evidence accumulation models of the sort depicted in Fig. 1C. Among other things, such studies demonstrate that people can adjust gradually the amount of evidence required for their responses, choosing deliberately among strategies ranging from relatively slow and accurate responses to relatively fast but error-prone ones—the so-called ‘‘speed–accuracy tradeoff’’ (SAT; e.g., Luce, 1986, Sec. 6.5; Pachella, 1974; Ruthruff, 1996). These adjustments are typically modeled as reflecting changes in a decision criterion like that shown in Fig. 1C, and the broad success of these models strongly supports the idea that evidence accumulation is a gradual process that can be terminated at any point (e.g., Ludwig & Davies, 2011). Even more direct evidence for graded conscious awareness comes from studies including various types of graded scales for indexing awareness (for a comparison of several methods, see, e.g., Sandberg, Timmermans, Overgaard, & Cleeremans, 2010). In some psychophysical studies, for example, people give graded ratings of their confidence in a judgment (e.g., Balakrishnan & Ratcliff, 1996) or of the clarity of the perceptual experience on which that judgment was based (e.g., Overgaard, Rote, Mouridsen, & Ramsøy, 2006; Ramsøy & Overgaard, 2004). Contrary to the idea that awareness is all-or-none, it is reported that people find graded awareness categories to be ‘‘both intuitive and easy to use’’ (Ramsøy & Overgaard, 2004, p. 10).2 Even more critically, objective judgment accuracy tends to increase gradually with people’s graded reports of their awareness (e.g., Balakrishnan & Ratcliff, 1996; Overgaard et al., 2006). Conversely, people have higher confidence in their correct judgments than in their errors, even when overall performance is near chance and people report that they are just guessing (e.g., Fleming, Weil, Nagy, Dolan, & Rees, 2010; Kunimoto, Miller, & Pashler, 2001). Such close correlations of graded reports and objective correctness are inconsistent with a system in which either perception or awareness varies in an all-or-none manner. A subtle yet telling phenomenon supporting graded awareness comes from ‘‘second guessing’’ experiments of the sort used to test high threshold theories of signal detection (e.g., Perez, Donoso, & Medina, 2010; Swets, Tanner, & Birdsall, 1961). In one classic study using this technique, for example, Bricker and Chapanis (1953) displayed letter strings very briefly in each trial and asked observers to indicate which of eight specific letter strings had been presented. Additionally, if the initial response was incorrect, observers were informed of that and asked to make further guesses. The key result was that observers’ subsequent guesses were above chance in trials where their initial responses had been incorrect. These abovechance guesses show that observers had extracted some partial information from stimuli that they could not correctly identify. The existence of such partial information is inconsistent with the idea of all-or-none stimulus perception, and it instead supports the idea that awareness of the stimulus accrues gradually. Analogous results in recognition memory experiments suggest that conscious recollection also varies in a graded manner under some conditions (e.g., Kellen & Klauer, 2011; Parks & Yonelinas, 2009). A vast body of quantitative data and models has also established that perceptual decisions under uncertainty can be modeled more accurately and parsimoniously under the assumption that the perceptual awareness of a stimulus varies in a graded manner, as is shown in Fig. 1C, rather than in an all-or-none fashion, as shown in Fig. 1B (e.g., Green & Swets, 1966; Macmillan & Creelman, 2005). Although detailed technical arguments about these models are beyond the scope of this article, the consensus of modeling work within the broad research areas of psychophysics and perception under uncertainty clearly favors graded theories over all-or-none theories—or, as they are often called, ‘‘threshold’’ theories—of perceptual awareness (Rouder & Morey, 2009). Finally, although a determined advocate of unconscious decision making in Libet-type paradigms might claim that none of the other evidence reviewed in this section is relevant to the spontaneous movement paradigm, it is noteworthy that there is even evidence of graded decision making within that paradigm. As was described earlier, most researchers have had participants indicate the times of their decisions using a Libet-type rotating spot and have concluded that participants were not aware of deciding to move until approximately 200 ms before they actually moved. Matsuhashi and Hallett (2008) used a much more elaborate model-based method of assessing decision times, however. Although the details of this method are outside the scope of the present analysis,3 the important point is that their analysis suggested participants were aware of deciding to move much earlier—almost 1.5 s before the movement. Thus, Matsuhashi and Hallett (2008)’s method supports the idea that awareness develops gradually over an extended time period even in the spontaneous movement task, as they suggest in their own account of their findings (see, e.g., their Fig. 1C). Indeed, the large difference between methods in measured awareness times is in itself difficult to explain in terms of a model with an abrupt all-or-none state transition as is shown in Fig. 1B.

2 Analogously, in some studies of binocular rivalry, people used a joystick to indicate which of two competing percepts was stronger (e.g., Fahle et al., 2011; Naber, Frässle, & Einhäuser, 2011). They often moved the joystick between its two extremes gradually over the course of approximately one s, suggesting a gradual conscious transition between the competing percepts. 3 In brief, Matsuhashi and Hallett (2008) presented tones at occasional random times during the spontaneous movement task, instructing participants to cancel an upcoming movement if a tone occurred when they were thinking about moving and to ignore any tone that occurred when they were not. Based on an analysis of the times between tones and movements, they concluded that participants must have been thinking about movements for as much as 1.42 s before they made them.

J. Miller, W. Schwarz / Consciousness and Cognition 24 (2014) 12–21

17

Although we have concentrated to this point on evidence of gradually-developing awareness of conscious decisions about sensory signals, it is easy to make the case that the same considerations would apply to conscious decisions about our own voluntary actions—as are assessed in Libet-type paradigms—because ‘‘voluntary action is therefore a form of decision making’’ (Haggard, 2008, p. 937). For one thing, there is no firm distinction between these two types of conscious awareness (i.e., sensory signals versus actions). Given the evidence for common internal representations underlying perception and action (e.g., Bortoletto, Mattingley, & Cunnington, 2011), it seems reasonable to believe that the internal states involved in perception are qualitatively similar to those associated with action-related decisions. Furthermore, the analysis of movementrelated physiological signals also provides strong evidence favoring a gradual view of action-related decisions (e.g., Gold & Shadlen, 2000; Gratton et al., 1988; Hanes & Schall, 1996; Schurger et al., 2012). Even ordinary language conveys quite clearly the graded nature of our own action-related decisions, as in phrases like ‘‘I am leaning toward doing X’’, ‘‘I am strongly considering doing X’’, and ‘‘I have almost decided to do X’’. As Haggard (2008, p. 942) puts it, ‘‘conscious intentions clearly come by degrees: one can be barely conscious that one is going to take the next step when walking, but intensely aware of pulling a trigger.’’. 3. A reinterpretation of brain activity preceding reports of conscious decisions Given the wide-ranging empirical support for the idea that conscious awareness of stimuli is graded, it seems reasonable to consider the possibility that a similarly graded process underlies participants’ decisions about when their conscious decisions take place in Libet-type spontaneous movement tasks (Fahle et al., 2011). At the start of each trial, it seems plausible that participants would already have a weak yet conscious urge to move within the next few seconds, simply because they know that their task is to make such movements (Klemm, 2010; Näätänen, 1985; Zhu, 2003). Based on the experimenter’s instructions, however, they would neither act immediately on this weak urge nor report that they were consciously aware of it. Instead, they might wait until the urge became strong enough to reach some preset criterion value. They would then act, and they would later report the moment at which the criterion was reached as W—the time when the decision to move was consciously finalized (Ringo, 1985). If the urge to make a specific movement develops gradually in accordance with this scenario, then evidence that brain processes precede W would not show that they precede any conscious awareness—just that they precede the point at which the criterion for acting and reporting awareness is reached. In that case, brain activity prior to W could well reflect sub-criterion conscious processing rather than unconscious processing. Indeed, the situation may be exactly as is illustrated in Fig. 1A and C, with nearly simultaneous onsets of decision-related brain activity and a conscious urge to make that decision, contrary to the view that an unconscious decision springs suddenly into consciousness. In this scenario, the fact that brain activity appears to emerge before the conscious decision—i.e., before W—is merely an artifact of the experimenter’s requirement that the observer impose an arbitrary criterion for making a binary judgment about an inherently gradual process that underlies decision making. Note that this challenge to the interpretation of W goes beyond the possibility of mere inaccuracy of measuring its timing, disputing instead the fundamental premise that W measures an abrupt underlying transition in the first place. As an analogy, consider a game of table tennis, played as a series of points, with the winner being the first to win the criterion number of 11. The discrete decision (i.e., determination of the winner) is not actually made until one player reaches 11, and yet it would be easy to find continuous signals related to the decision outcome before that point (e.g., the players’ facial expressions). These preceding signals are not the real causes of the decision, nor do they indicate that the decision itself (i.e., the 11th point) is reached without conscious awareness. Instead, these preceding signals are merely correlates of the graded states (i.e., the intermediate results) leading up to the discrete decision. 4. General discussion Our main thesis—Libet-type studies do not unequivocally show that decision-related brain activity precedes conscious awareness—depends crucially on the possibility that conscious awareness could vary gradually. We have relied mainly on theoretical precedents and empirical results within psychology and the neurosciences to establish that gradual awareness is not only possible but indeed quite plausible. As noted earlier, similar arguments against the all-or-none view of consciousness have also recently been made by Matsuhashi and Hallett (2008) based on an alternative measure of conscious decision times and by Fahle et al. (2011) within the context of experiments on binocular rivalry. Mele (2006, 2007, 2010) has recently presented philosophical arguments for a position that shares with ours the emphasis on a gradual decision-making process but differs from ours in the extent to which this process is conscious. In Mele (2006, p. 190), for example, he wrote: In saying that deciding is momentary, I mean to distinguish it from, for example, a combination of deliberating and deciding. Someone who is speaking loosely may say ‘I was up all night deciding to sell my house’ when what he means is that he was up all night deliberating or fretting about whether to sell his house and eventually decided to sell it. Deciding to A, on my view, is not a process but a momentary mental action of forming an intention to A, ‘form’ being understood as an action verb.

18

J. Miller, W. Schwarz / Consciousness and Cognition 24 (2014) 12–21

In this passage, Mele clearly conceives of a gradual decision-making process of the sort depicted in Fig. 1C, which he calls ‘‘deliberation’’. A minor difference between Mele’s position and ours is that he reserves the term ‘‘decision’’ for the final resolution at the end of that process, using the term ‘‘urges’’ for the precursors of the decisions (i.e., the sub-criterion levels of activation in Fig. 1C). In contrast, we regard the criterion-crossing as an otherwise unexceptional event within the decisionmaking process and hence prefer to use ‘‘decision’’ as short-hand for the entire process rather than reserving it for the final termination. The major difference between our position and Mele’s, however, is that he accepts the unconscious nature of the gradual process preceding the decision: ‘‘rather than. . .making a decision of which he is not conscious, the agent instead acquires an urge. . .of which he is not conscious’’ (p. 245, italics in original). In short, Mele suggests that Libet et al.’s early readiness potential may reflect a subconscious urge rather than a subconscious decision. In contrast, we think the gradually developing activations may perfectly well be conscious, even if they are too weak to reach the participant’s criterion for reporting that the decision has been made. It should be emphasized that, with regard to the interpretation of Libet-type studies, the important issue is whether people are unconscious of the gradual precursors of the decision, not whether the decision itself is made abruptly or gradually. On the basis of the evidence reviewed in Section 2, our position is that people very likely do have some awareness of the gradual changes in evidence accumulation underlying their decisions. Indeed, we think Mele’s house-selling scenario is a perfect example in support of our position, because it is easy to imagine that the homeowner is completely conscious of gradual fluctuations in the urge to sell even before reaching the ultimate decision about whether to do so. Schurger et al. (2012) have recently suggested another criticism of Libet-type studies that shares certain features with our approach but is nonetheless fundamentally different. In brief, Schurger et al. (2012) argued that RP onset may be too early to use as a neural marker of the decision, whereas we argue that W may be too late to use as a marker for conscious awareness of the decision. More specifically, based on EEG data and a theoretical model, Schurger et al. (2012) objected that the gradually-rising early RP may not reflect the neural decision as Libet claimed. Instead, it may reflect ‘‘spontaneous fluctuations of neural firing rate which at some other time have not preceded a movement’’ (Schurger et al., 2012, p. 7), with the actual neural decision happening later, at the point when the amount of neural activity crosses some threshold. On their view, then, Libet et al. (1983) were not justified in concluding from early RP onset that the neural decision preceded W, because the RP may actually have started well before the neural decision was made. Crucially, according to Schurger et al. (2012), their ‘‘study is not concerned with the subjective urge to move’’ on which we have focused, and the neural activity they discussed was ‘‘conceptually distinct from the conscious decision to move’’ (p. E2905) that we emphasize may also be gradual. Thus, we join with Schurger et al. in disputing Libet’s conclusion that the neural decision takes place before awareness of that decision, but we do so for a different reason. In essence, Schurger et al. (2012) argued that the neural decision might take place much later than Libet et al. (1983) claimed, because RP onset could precede that final neural decision. In contrast, we say that awareness could begin much earlier than Libet et al. (1983) claimed, because the earliest levels of partial awareness could precede W reports (cf. Fig. 1C). It appears that Libet et al. (1983) were not completely unaware of the possibility that gradual rather than abrupt processes might underlie decision making in their task. They only acknowledged the possibility of gradual transitions at the neural level, however, writing ‘‘it might be proposed that neural activities. . .must achieve some threshold. . .before the brain ‘decides’ to act’’ (p. 637). They dismissed this idea as unimportant—‘‘such a [gradual] model [of neural activity] would not fundamentally affect our conclusion’’ (p. 637)—arguing that the possibility of a gradual neural decision-making process presents no fundamental threat to their argument. On the other hand, they did not address the possibility that conscious awareness might develop gradually, as we have emphasized here, apparently not recognizing that such a gradual model—together with a threshold for reporting awareness—could explain their results without positing any unconscious development of intention. It is possible that experimenters could at least partially overcome the problem of gradual conscious awareness by instructing observers to use a low criterion, reporting as W the first moment at which they had any awareness of an urge to act. Such W reports might then reflect the earliest onset of conscious awareness, and evidence of decision-related brain activity preceding such reports could not be attributed to sub-criterion levels of consciousness. In practice, however, it might be quite difficult for researchers to formulate instructions that would appropriately encourage participants to use a low criterion, because such instructions are somewhat at odds with the basic demand characteristics of the experimental paradigms used in this research area. In the experiment of Libet et al. (1983), for example, participants were asked to generate occasional spontaneous movements and then to report the moment of being consciously aware of having decided to move. Suppose that the experimenter emphasized that the participant should report the earliest awareness of having any conscious intention to move. A participant might legitimately report, ‘‘I first decided to move at the beginning of the experiment, when I was told the instructions. That was the first point at which I was consciously aware that I would move in this trial—and in each of the other trials as well.’’ Such a report might describe accurately the participant’s overall experience with the experiment, so encouraging participants to use a very low awareness criterion might result in such unwanted reports. In the actual paradigm of Libet et al. (1983), of course, participants would never have made such reports because of demands implicit in the experimental setup. For example, the experimenter asked for the time reports to be made using the rotating clock, which constrained the reports to fall within the time interval during which the clock was present and rotating. Moreover, the experimenter emphasized that the movements should be spontaneous rather than preplanned, which clearly indicates that the reported moment should be just before the movement itself. Interestingly, Libet et al. (1983) acknowledged that participants might sometimes be aware of what might be called ‘‘a looser preintentionality or

J. Miller, W. Schwarz / Consciousness and Cognition 24 (2014) 12–21

19

general preparation-to-act-soon’’ (Libet, 1985, p. 532), and their analysis excluded trials in which participants reported such preplanning. Most likely, though, the additional requirement for participants to distinguish between trials with versus without preplanning established the implicit expectation that some trials should be reported as having no preplanning. In a model with graded development of conscious awareness, the effect of such instructions would simply be to increase the criterion level at which participants reported that they made their decisions to act, so that some movements could be reported as unplanned. Our fundamental thesis—that conscious awareness could build gradually as shown in Fig. 1C rather than abruptly as shown in Fig. 1B—has also been used to challenge many studies that have been claimed to demonstrate perception without awareness (e.g., Cheesman & Merikle, 1986; Eriksen, 1956; Haase & Fisk, 2004). Many such studies support subliminal perception only ‘‘to the degree that an inappropriate, threshold model [of awareness] guides theorizing and experimentation’’ (Macmillan & Creelman, 2005, p. 259). One classic example is the study of Lazarus and McCleary (1951), who found that galvanic skin responses (GSRs) discriminated between shock-associated and control letter strings even in trials where participants incorrectly named these strings. This was interpreted as ‘‘autonomic discrimination without awareness’’ (Lazarus & McCleary, 1951, p. 113), based on the assumption that participants had no awareness of letter strings that they could not correctly report (for an analogous example, see Libet, Alberts, Wright, & Feinstein, 1967). As was mentioned earlier, however, Bricker and Chapanis (1953) showed that this assumption is clearly wrong (for analogous recent results, see Parks & Yonelinas, 2009). Participants made better-than-chance second guesses even when their first identification responses were incorrect, showing that they must have conscious access to partial information about a stimulus that is below the threshold for correct identification. Our suggestion is that early decision-related brain activity in Libet-type paradigms could reflect subcriterion levels of awareness in an exactly analogous manner. There appear to be at least three possible methods of taking gradual awareness into account within studies of the relation between consciousness and brain activity. One possibility is to model explicitly the gradual development of conscious awareness depicted in Fig. 1C. By manipulating the criterion level systematically across conditions, for example, it might be possible to trace the gradual development of a decision, much as speed–accuracy tradeoff studies trace the accumulation of discriminative information (e.g., Ludwig & Davies, 2011). Clearly, the relation between brain activity and conscious awareness could be characterized more accurately using this more nuanced measure of consciousness. A second and related possibility is to develop new methods by which decisions can be measured gradually rather than discretely, for example by manipulating instructions about when to report conscious awareness or by using continuous response devices or by using interruption-based measurement procedures (e.g., Matsuhashi & Hallett, 2008; Schurger et al., 2012). Despite recent major advances in the measurement of brain processes, there have been few if any corresponding advances in measures of conscious states, and interpretations of Libet-type paradigms are clearly most severely limited by the weakness of the latter measures. Attempts in this direction have been already made within perceptual discrimination tasks (e.g., Anderson, Chiu, Huette, & Spivey, 2011; Fahle et al., 2011; Graziano, Polosecki, Shalom, & Sigman, 2011; Lachter, Johnston, Corrado, & McClelland, 2009; Naber et al., 2011; Resulaj, Kiani, Wolpert, & Shadlen, 2009; Spivey et al., 2005; for a review, see Freeman et al., 2011; for cautions, see Van der Wel et al., 2009) Van der Wel, Eder, Mitchel, Walsh, and Rosenbaum (e.g., Lau et al., 2007). Such experimental manipulations could allow more direct titration of the gradual relationship between brain states and consciousness consistent with the more general program of neurophenomenology (e.g., Varela, 1996). In the final analysis, the possibility of gradually-developing conscious awareness seriously undermines Libet-type arguments that decisions are made unconsciously. On the gradual view, the observed early brain activity need not precede any conscious awareness of decision-making—but only the attainment of some possibly rather high criterion for reporting that the decision has been made. Naturally, by undermining the argument for unconscious decision making, the possibility of gradual conscious awareness also invalidates the conclusion that consciousness is epiphenomenal—and its corollaries regarding free will and moral responsibility—that these neurobiological arguments seemed to imply. References Anderson, S. E., Chiu, E., Huette, S., & Spivey, M. J. (2011). On the temporal dynamics of language-mediated vision and vision-mediated language. Acta Psychologica, 137, 181–189. Baars, B. J., & Gage, N. M. (2010). Cognition, brain, and consciousness (2nd ed.). Amsterdam: Elsevier. Balakrishnan, J. D., & Ratcliff, R. (1996). Testing models of decision making using confidence ratings in classification. Journal of Experimental Psychology: Human Perception & Performance, 22, 615–633. Bogacz, R. (2007). Optimal decision-making theories: Linking neurobiology with behaviour. Trends in Cognitive Sciences, 11, 118–125. Bortoletto, M., Mattingley, J. B., & Cunnington, R. (2011). Action intentions modulate visual processing during action perception. Neuropsychologia, 49, 2097–2104. Breitmeyer, B. G. (1985). Problems with the psychophysics of intention. Behavioral & Brain Sciences, 8, 539–540. Bricker, P. D., & Chapanis, A. (1953). Do incorrectly perceived tachistoscopic stimuli convey some information? Psychological Review, 60, 181–188. Brown, S. D., & Heathcote, A. (2008). The simplest complete model of choice reaction time: Linear ballistic accumulation. Cognitive Psychology, 57, 153–178. Cheesman, J., & Merikle, P. M. (1986). Distinguishing conscious from unconscious perceptual processes. Canadian Journal of Psychology, 40, 343–367. Corallo, G., Sackur, J., Dehaene, S., & Sigman, M. (2008). Limits on introspection: Distorted subjective time during the dual-task bottleneck. Psychological Science, 19, 1110–1117. Dehaene, S. (2001). The cognitive neuroscience of consciousness. Cambridge, MA: MIT Press. Dehaene, S., & Changeux, J. P. (2011). Experimental and theoretical approaches to conscious processing. Neuron, 70, 200–227. Dehaene, S., & Naccache, L. (2001). Towards a cognitive neuroscience of consciousness: Basic evidence and a workspace framework. Cognition, 79, 1–37. Eriksen, C. W. (1956). Subception: Fact or artifact? Psychological Review, 63, 74–80. Fahle, M. W., Stemmler, T., & Spang, K. M. (2011). How much of the ‘‘unconscious’’ is just pre-threshold? Frontiers in Human Neuroscience, 5, 1–6.

20

J. Miller, W. Schwarz / Consciousness and Cognition 24 (2014) 12–21

Filevich, E., Kühn, S., & Haggard, P. (2013). There is no free won’t: Antecedent brain activity predicts decisions to inhibit. PLoS One, 8, e53053. Fleming, S. M., Weil, R. S., Nagy, Z., Dolan, R. J., & Rees, G. (2010). Relating introspective accuracy to individual differences in brain structure. Science, 329, 1541–1543. Freeman, J. B., Dale, R., & Farmer, T. A. (2011). Hand in motion reveals mind in motion. Frontiers in Psychology, 2, 1–6. Fried, I., Mukamel, R., & Kreiman, G. (2011). Internally generated preactivation of single neurons in human medial frontal cortex predicts volition. Neuron, 69, 548–562. Gold, J. I., & Shadlen, M. N. (2000). Representation of a perceptual decision in developing oculomotor commands. Nature, 404, 390–394. Gold, J. I., & Shadlen, M. N. (2007). The neural basis of decision making. Annual Review of Neuroscience, 30, 535–574. Gratton, G., Coles, M. G. H., Sirevaag, E., Eriksen, C. W., & Donchin, E. (1988). Pre- and post-stimulus activation of response channels: A psychophysiological analysis. Journal of Experimental Psychology: Human Perception & Performance, 14, 331–344. Graziano, M., Polosecki, P., Shalom, D. E., & Sigman, M. (2011). Parsing a perceptual decision into a sequence of moments of thought. Frontiers in Integrative Neuroscience, 5, 1–8. Green, D. M., & Swets, J. A. (1966). Signal detection theory and psychophysics. New York, NY, US: Wiley. Gregory, R. L. (2004). The Oxford companion to the mind (2nd ed.). Oxford, England: Oxford University Press. Grill-Spector, K., Kushnir, T., Hendler, T., & Malach, R. (2000). The dynamics of object-selective activation correlate with recognition performance in humans. Nature Neuroscience, 3, 837–843. Gurwitsch, A. (1964). The field of consciousness (Vol. 2). Pittsburgh, PA: Duquesne University Press. Haase, S. J., & Fisk, G. D. (2004). Valid distinctions between conscious and unconscious perception? Perception & Psychophysics, 66, 868–871. Haggard, P. (1999). Perceived timing of self-initiated actions. In G. Aschersleben, T. Bachmann, & J. Müsseler (Eds.), Cognitive contributions to the perception of spatial and temporal events. Advances in psychology (Vol. 129, pp. 215–231). Amsterdam, Netherlands: North Holland. Haggard, P. (2005). Conscious intention and motor cognition. Trends in Cognitive Sciences, 9, 290–295. Haggard, P. (2008). Human volition: Towards a neuroscience of free will. Nature Reviews Neuroscience, 9, 934–946. Haggard, P., & Eimer, M. (1999). On the relation between brain potentials and the awareness of voluntary movements. Experimental Brain Research, 126, 128–133. Hanes, D. P., & Schall, J. D. (1996). Neural control of voluntary movement initiation. Science, 274, 427–430. Husserl, E. (2001). Logical investigations (Vol. 1). Routledge. Joordens, S., Spalek, T. M., Razmy, S., & Van Duijn, M. (2004). A clockwork orange: Compensation opposing momentum in memory for location. Memory & Cognition, 32, 39–50. Joordens, S., Van Duijn, M., & Spalek, T. M. (2002). When timing the mind one should also mind the timing: Biases in the measurement of voluntary actions. Consciousness & Cognition, 11, 231–240. Kellen, D., & Klauer, K. C. (2011). Evaluating models of recognition memory using first- and second-choice responses. Journal of Mathematical Psychology, 55, 251–266. Klemm, W. R. (2010). Free will debates: Simple experiments are not so simple. Advances in Cognitive Psychology, 6, 47–65. Kunimoto, C., Miller, J. O., & Pashler, H. E. (2001). Confidence and accuracy of near-threshold discrimination responses. Consciousness & Cognition, 10, 294–340. Lachter, J., Johnston, J. C., Corrado, G. S., & McClelland, J. L. (2009). Toward online measurement of decision state. Poster presented at the 50th annual meeting of the psychonomic society, Boston, Massachusetts, November. Lau, H. C., Rogers, R. D., & Passingham, R. E. (2007). Manipulating the experienced onset of intention after action execution. Journal of Cognitive Neuroscience, 19, 81–90. Lavazza, A., & De Caro, M. (2010). Not so fast. On some bold neuroscientific claims concerning human agency. Neuroethics, 3, 23–41. Lazarus, R., & McCleary, R. (1951). Autonomic discrimination without awareness: A study of subception. Psychological Review, 58, 113–122. Leuthold, H., Sommer, W., & Ulrich, R. (2004). Preparing for action: Inferences from CNV and LRP. Journal of Psychophysiology, 18, 77–88. Libet, B. (1985). Unconscious cerebral initiative and the role of conscious will in voluntary action. Behavioral & Brain Sciences, 8, 529–566. Libet, B. (2003). Timing of conscious experience: Reply to the 2002 commentaries on Libet’s findings. Consciousness & Cognition, 12, 321–331. Libet, B., Alberts, W. W., Wright, E. W., Jr., & Feinstein, B. (1967). Responses of human somato-sensory cortex to stimuli below the threshold for conscious sensation. Science, 158, 1597–1600. Libet, B., Gleason, C. A., Wright, E. W., Jr., & Pearl, D. K. (1983). Time of conscious intention to act in relation to onset of cerebral activity (readinesspotential): The unconscious initiation of a freely voluntary act. Brain, 106, 623–642. Luce, R. D. (1986). Response times: Their role in inferring elementary mental organization. Oxford, England: Oxford University Press. Ludwig, C. J. H., & Davies, J. R. (2011). Estimating the growth of internal evidence guiding perceptual decisions. Cognitive Psychology, 63, 61–92. Macmillan, N. A., & Creelman, C. D. (2005). Detection theory: A user’s guide (2nd ed.). Hillsdale, NJ: Erlbaum. Mangan, B. (1993). Taking phenomenology seriously: The fringe and its implications for cognitive research. Consciousness and Cognition, 2, 89–108. Matsuhashi, M., & Hallett, M. (2008). The timing of the conscious intention to move. European Journal of Neuroscience, 28, 2344–2351. Mele, A. R. (2006). Free will: Theories, analysis, and data. In S. Pockett & W. P. Banks (Eds.), Does consciousness cause behavior? (pp 187–205). Cambridge, MA, US: MIT Press. Mele, A. R. (2007). Decisions, intentions, urges, and free will: Why Libet has not shown what he says he has. In J. K. Campbell, M. O’Rourke, & H. Silverstein (Eds.), Causation and explanation (pp. 241–264). Cambridge, MA, US: MIT Press. Mele, A. R. (2009). Effective intentions: The power of conscious will. Oxford, England: Oxford University Press. Mele, A. R. (2010). Conscious deciding and the science of free will. In R. F. Baumeister, A. R. Mele, & K. D. Vohs (Eds.), Free will and consciousness, How might they work? (pp 43–65). Oxford, England: Oxford University Press. Metzinger, T. (Ed.). (2000). Neural correlates of consciousness: Empirical and conceptual questions. Cambridge, MA, US: MIT Press. Miller, J. O., Shepherdson, P., & Trevena, J. (2011). Effects of clock monitoring on electroencephalographic activity: Is unconscious movement initiation an artifact of the clock? Psychological Science, 22, 103–109. Miller, J. O., Vieweg, P., Kruize, N., & McLea, B. (2010). Subjective reports of stimulus, response, and decision times in speeded tasks: How accurate are decision time reports? Consciousness & Cognition, 19, 1013–1036. Milner, A. D., & Rugg, M. D. (1992). The neuropsychology of consciousness. San Diego, CA, US: Academic Press. Näätänen, R. (1985). Brain physiology and the unconscious initiation of movements. Behavioral & Brain Sciences, 8, 549. Naber, M., Frässle, S., & Einhäuser, W. (2011). Perceptual rivalry: Reflexes reveal the gradual nature of visual awareness. PLoS One, 6, e20910. Obhi, S. S., & Haggard, P. (2004). Free will and free won’t. American Scientist, 92, 358–365. Overgaard, M., Rote, J., Mouridsen, K., & Ramsøy, T. Z. (2006). Is conscious perception gradual or dichotomous? A comparison of report methodologies during a visual task. Consciousness & Cognition, 15, 700–708. Pachella, R. G. (1974). The interpretation of reaction time in information-processing research. In B. H. Kantowitz (Ed.), Human information processing: Tutorials in performance and cognition (pp. 41–82). Hillsdale, NJ, US: Erlbaum. Parks, C. M., & Yonelinas, A. P. (2009). Evidence for a memory threshold in second-choice recognition memory responses. Proceedings of the National Academy of Sciences of the United States of the America, 106, 11515–11519. Perez, C. A., Donoso, J. R., & Medina, L. E. (2010). A critical experimental study of the classical tactile threshold theory. BMC Neuroscience, 11, 1–8. Pockett, S. (2004). Does consciousness cause behaviour? Journal of Consciousness Studies, 11, 23–40. Pockett, S., & Miller, A. (2007). The rotating spot method of timing subjective events. Consciousness & Cognition, 16, 241–254.

J. Miller, W. Schwarz / Consciousness and Cognition 24 (2014) 12–21

21

Purcell, B. A., Heitz, R. P., Cohen, J. Y., Schall, J. D., Logan, G. D., & Palmeri, T. J. (2010). Neurally constrained modeling of perceptual decision making. Psychological Review, 117, 1113–1143. Radder, H., & Meynen, G. (2013). Does the brain ‘‘initiate’’ freely willed processes? A philosophy of science critique of Libet-type experiments and their interpretation. Theory & Psychology, 23, 3–21. Ramsøy, T. Z., & Overgaard, M. (2004). Introspection and subliminal perception. Phenomenology and the Cognitive Sciences, 3, 1–23. Ratcliff, R. (2006). Modeling response signal and response time data. Cognitive Psychology, 53, 195–237. Ratcliff, R., Cherian, A., & Segraves, M. (2003). A comparison of macaque behavior and superior colliculus neuronal activity to predictions from models of two-choice decision making. Journal of Neurophysiology, 90, 1392–1407. Rees, G., Kreiman, G., & Koch, C. (2002). Neural correlates of consciousness in humans. Nature Reviews Neuroscience, 3, 261–270. Resulaj, A., Kiani, R., Wolpert, D. M., & Shadlen, M. N. (2009). Changes of mind in decision-making. Nature, 461, 263–266. Ringo, J. L. (1985). Timing volition: Questions of what and when about W. Behavioral & Brain Sciences, 8, 550–551. Roskies, A. L. (2006). Neuroscientific challenges to free will and responsibility. Trends in Cognitive Sciences, 10, 419–423. Roskies, A. L. (2010). Freedom, neural mechanism, and consciousness. In R. F. Baumeister, A. R. Mele, & K. D. Vohs (Eds.), Free will and consciousness, How might they work? (pp 153–171). Oxford, England: Oxford University Press. Rouder, J. N., & Morey, R. D. (2009). The nature of psychological thresholds. Psychological Review, 116, 655–660. Ruthruff, E. D. (1996). A test of the deadline model for speed–accuracy tradeoffs. Perception & Psychophysics, 58, 56–64. Sandberg, K., Timmermans, B., Overgaard, M., & Cleeremans, A. (2010). Measuring consciousness: Is one measure better than the other? Consciousness & cognition, 19, 1069–1078. Schall, J. D. (2001). Neural basis of deciding, choosing and acting. Nature Reviews Neuroscience, 2, 33–42. Schall, J.D. (2003). Neural selection and control of action. Chapter for A&P 20. Schurger, A., Sitt, J. D., & Dehaene, S. (2012). An accumulator model for spontaneous neural activity prior to self-initiated movement. Proceedings of the National Academy of Sciences of the United States of the America, 109, E2904–E2913. Searle, J. R. (2000). Consciousness, free action and the brain. Journal of Consciousness Studies, 7, 3–22. Seth, A. K., Dienes, Z., Cleeremans, A., Overgaard, M., & Pessoa, L. (2008). Measuring consciousness: Relating behavioral and neurophysiological approaches. Trends in Cognitive Sciences, 12, 314–321. Smith, P. L. (2000). Stochastic dynamic models of response time and accuracy: A foundational primer. Journal of Mathematical Psychology, 44, 408–463. Smith, K. (2011). Neuroscience vs philosophy: Taking aim at free will. Nature, 477, 23–25. Smith, P. L., & Ratcliff, R. (2009). An integrated theory of attention and decision making in visual signal detection. Psychological Review, 116, 283–317. Soon, C. S., Brass, M., Heinze, H. J., & Haynes, J. D. (2008). Unconscious determinants of free decisions in the human brain. Nature Neuroscience, 11, 543–545. Spivey, M. J., Grosjean, M., & Knoblich, G. (2005). Continuous attraction toward phonological competitors: Thinking with your hands. Proceedings of the National Academy of Sciences of the United States of the America, 102, 10393–10398. Swets, J. A., Tanner, W. P., Jr., & Birdsall, T. G. (1961). Decision processes in perception. Psychological Review, 68, 301–340. Trevena, J. A., & Miller, J. O. (2002). Cortical movement preparation before and after a conscious decision to move. Consciousness & Cognition, 11, 162–190. Trevena, J. A., & Miller, J. O. (2013). Brain preparation before a voluntary action: Evidence against unconscious movement initiation. Consciousness & Cognition, 19, 447–456. Usher, M., & McClelland, J. L. (2001). The time course of perceptual choice: The leaky, competing accumulator model. Psychological Review, 108, 550–592. Van de Grind, W. (2002). Physical, neural, and mental timing. Consciousness & Cognition, 11, 241–264. Van der Wel, R. P. R. D., Eder, J. R., Mitchel, A. D., Walsh, M. M., & Rosenbaum, D. A. (2009). Trajectories emerging from discrete versus continuous processing models in phonological competitor tasks: A commentary on Spivey, Grosjean, and Knoblich (2005). Journal of Experimental Psychology: Human Perception & Performance, 35, 588–594. Van Gaal, S., & Lamme, V. A. F. (2012). Unconscious high-level information processing: Implication for neurobiological theories of consciousness. The Neuroscientist, 18, 287–301. Varela, F. J. (1996). Neurophenomenology: A methodological remedy for the hard problem. Journal of Consciousness Studies, 3, 330–349. Wegner, D. M. (2002). The illusion of conscious will. Cambridge, MA: MIT Press. Zhu, J. (2003). Reclaiming volition: An alternative interpretation of Libet’s experiment. Journal of Consciousness Studies, 10, 61–77.