Neurodynamics of time consciousness: An extensionalist explanation of apparent motion and the specious present via reentrant oscillatory multiplexing

Highlights

• Original extensionalist view fully situated in philosophy and neuroscience.

• Differential latency model of apparent motion provides extensionalist explanation.

• Unified view of conscious continuity and discreteness given by phase multiplexing.

• Oscillatory model provides a unifying framework for the specious present.

• This model appears naturally consonant with predictive processing theories.

Abstract

This paper situates an original model of reentrant oscillatory multiplexing within the philosophy of time consciousness to argue for an extensionalist theory of the specious present. I develop a detailed differential latency model of apparent motion to show how the ordinality of experiential content is isomorphic to the ordinality of relevant brain processes. I argue that the theory presented has resources to account for other key features of the specious present, including the representational discreteness between successive conscious moments as well as the phenomenological continuity between them. This work not only shows the plausibility of an extensionalist philosophical theory, it also illustrates the utility of differential latency views in squaring temporal illusions with empirically supported neurodynamics.

Introduction

I will be focusing on a debate in the philosophy of temporal consciousness about the nature of immanent temporal consciousness – i.e., the subjective experience of “now” – what James (1890) referred to as the “specious present,” the “short duration of which we are immediately and incessantly sensible” (p. 632; Andersen, 2014). The specious present is remarkable in that it includes both a perception of change or motion (succession) and the experience of conscious unity. Consider the oft-cited phenomenology of seeing a shooting star: even though motion and change are experienced as the meteor briefly falls, the entire event is experienced in one perceptual gulp, as a single conscious moment. Although the specious present has been characterized by a wide variety of durations, the focus here is the subjective duration of a maximally unified phenomenological experience. Hence, my goal is not to provide a universal theory of time consciousness, nor to account for all temporal phenomena (cf. Viera, 2019); rather, my claims are restricted to the specious present, characterized by the experience of conscious unity amidst brief phenomenological change.²

There are two main views in the debate (Dainton, 2010, Grush, 2007, Hoerl, 2009, Lee, 2014b), and they both agree that the representational content of our immediate experience presents a temporal interval, but they disagree over the nature of the relationship between that content and the processes that generate it.³ There are various ways to characterize and label these rival views⁴, but the basic dichotomy in which I’m interested concerns whether the temporal properties experienced during the specious present are, or are not, explained by resemblance with the temporal properties of experiences. Extensionalists, on the one hand, explain the features of experienced time through resemblance to the temporal features of conscious experiences (Dainton, 2010, Hoerl, 2009, Phillips, 2014, Stern, 1897). On this view, the order and temporal extension of conscious contents is explained by the order and temporal extension of experiential processes. Atomists, on the other hand, do not think the timing of experiential processes explains the temporal content represented in experiences (Dennett and Kinsbourne, 1992, Grush, 2007, Lee, 2014a, Lee, 2014b). Collectively, they hold that experiential processes are much briefer than the temporal interval experienced and can also invert ordinality, thus violating various kinds of resemblance.

Hence, the central question in this literature is whether or not the immediate experience of time is, or is not, explained by the temporal nature of experience itself. Geoffrey Lee (2014b) usefully describes the different views this way: “I define atomism as the view that temporal experiences are never process-like…[and] do not contain shorter experiences as temporal parts…[while] the idea of the [extensionalist] view is that experiential stages of temporal experiences are distinct property instantiations – for example, they are realized by different physical events happening at different times” (p. 152–3). Below, I will introduce a neurodynamical model of the specious present that is process-like, involving the integration of information over time via reentrant oscillatory multiplexing.

Although philosophers have tended to describe the debate in brain-neutral terms, I think the most perspicuous way to unpack the issue is around the distinction between brain time (i.e., the timing of neural processes) and experienced time (i.e,. time as consciously experienced).⁵ Extensionalists hold there is an explanatory resemblance between the two (brain time -> experienced time); atomists deny that any such resemblance, even if it existed, would be explanatory. Putting the debate this way helps reveal how developments in the neurosciences dovetail with sometimes insular philosophical discussions on temporal consciousness. Specifically, neuroscientific differential latency views – which hold that apparent mismatches in brain time and experienced time can be harmonized by taking account of the fact that processing latencies depend on stimulus features, task conditions, and attentive focus, etc. (Breitmeyer and Ganz, 1976, Purushothaman et al., 1998, Whitney and Murakami, 1998; Patel et al., 2000, Bedell et al., 2003, Ogmen et al., 2004, Baldo and Caticha, 2005, Bachmann, 2013, Wutz and Melcher, 2014) – can be appreciated as natural allies of the extensionalist position because they leverage latency variance to maintain resemblance of brain and experienced time. Though this alliance isn’t theoretically necessary, I will be presenting a theory that honors both differential latency and extensionalism.

Atomist-type views have attracted many adherents (Dennett and Kinsbourne, 1992, Gallagher, 2017, Grush, 2007, Husserl, 2014, Lee, 2014a, Lee, 2014b, Molder, 2014). There are three main reasons. First, in general, representational contents don’t resemble and aren’t resembled by the properties of representing processes. Consciously representing something as red or heavy doesn’t involve, nor would be explained by, brain processes becoming red or heavy, for example. Prima facie, the same could be assumed for temporal representation: i.e., that represented time isn’t explained through resemblance with neurodynamical processing time. Second, many find it intuitive, and even neurodynamically likely, to think of the neural processing that generates temporal experience as involving brief signal convergence culminating in discrete integration events (Gallagher and Zahavi, 2014, Lee, 2014a). The insularity of these postulated (sequences of) integration events motivates an atomist picture. The third main reason is the existence of temporal illusions, cases in which experienced time and the timing of experiences seem to diverge: prima facie, a view espousing non-isomorphism between the contents of experience and the timing of experiential processes sits more comfortably with these data. The chief bone of contention here is the apparent motion illusion: proponents of all the major philosophical views have weighed in on it (e.g., Arstila, 2016a, Dainton, 2008, Dainton, 2010, Grush, 2007), as have many neuroscientists (e.g., Adelsen and Bergen, 1985, Grossberg and Rudd, 1989, Herzog and Ogmen, 2015, Jantzen et al., 2012, Sanders et al., 2014, VanRullen, 2016).

By developing an empirically informed neuroscience of apparent motion, highlighting mechanisms like reentrant circuitry and oscillatory multiplexing that have not generally made it onto philosophers’ radars to date in this debate, I will show that extensionalists have nothing to fear from apparent motion – and, by extension, other temporal illusions. An in-depth understanding of the relevant neurodynamics shows that the ordinal phenomenology of apparent motion resembles and hence is arguably explained by the ordinal progression of oscillatory mechanisms.⁶ I thereby endeavor to show that extensionalism is a viable and plausible theory of immanent temporal consciousness – i.e, the specious present (cf. Arstila & Lloyd, 2014) – despite temporal illusions and the fact that representational content determination is typically non-resemblant vis-à-vis its neural vehicles.

It is important to be very clear about what I will be claiming. Although I endorse an extensionalist theory of the specious present, I will not be trying to resolve the larger, more general, extensionalism-atomism debate on temporal consciousness. This is for reasons of space and methodological concern. Two concerns are foremost. One, I am doubtful that a single theory of temporal consciousness can account for the host of very disparate temporal experiences we can have (cf. Grush, 2015, Lloyd and Arstila, 2014, Montemayor, 2017, Viera, 2019). Hence, I restrict my analysis to the specious present as previously characterized. Two, Hoerl (2017) convincingly points out some serious obstacles to distinguishing between metaphysical views via empirical inquiry. Consider that while extensionalists highlight the temporal flow, extension and continuity of representational vehicles to explain the flow, extension and continuity of experiential content, some atomists have claimed they can posit a continuum of overlapping representational vehicles to account for the same phenomena (Gallagher and Zahavi, 2014, Lee, 2014b). However, if that is the case, then it is unclear (at best) what empirically accessible differences might exist between the kind of extensionalism I advocate and (what we might call) an extended overlap atomism. While I think it is questionable whether the atomist can legitimately claim such features of the view without contradiction (cf. White, 2018), I do not have adequate space for that discussion.⁷ Hence, my goal is to show the consonance of the reentrant oscillatory multiplexing (ROM) neurodynamical model with an extensionalist view of the specious present. I will not attempt to show that the ROM model is incompatible with atomist views. But there is still a large dialectical victory to be won! Since oscillatory approaches are typically thought to imply anti-extensionalist theories of temporal experience (cf. Busch and VanRullen, 2014, Metzinger, 1995, Ruhnau, 1995), if I can demonstrate the natural coherence of my model and extensionalism, I will have achieved my aim. With that in mind, a schematic overview of the intended integrative dialogue is presented in Fig. 1.

As mentioned, my defense of extensionalism is an extension of differential latency theories. Differential latency views show that stimulus timing (an objective external property) and the timing of neural processes (an objective internal property) can and do come apart. While stimulus timing and experienced temporal order can be asymmetrical, differential latency views show how neurodynamical progression resembles the ordinality of experienced content. Baldo and Caticha (2005) provide a clear description of the strategy I will employ:

The existence of a sequence of processing steps naturally includes intrinsic temporal delays…[that] can influence not only the magnitude of the [temporal illusion] but the very nature of the perceptual effect… This conclusion lies at the very core of the differential latencies account.…a stimulus delivered to the input layer after the presentation of a previous stimulation could be able to catch up with the ongoing neural activity produced by the preceding stimulus and modify it before its perceptual actualization (2627-8).

Work on the flash-lag illusion and color-motion asynchronies has already demonstrated the fecundity of the differential latency view (Bachmann, 2013, Bedell et al., 2003, Bedell et al., 2006, Kafaligonul et al., 2010, Ogmen et al., 2004, Patel et al., 2000, Purushothaman et al., 1998). The present work attempts to add three features to the discussion: (1) to more deeply integrate the philosophical work on temporal consciousness with the neuroscientific approaches already on the table (see Muller & Nobre, 2014); (2) to provide a more detailed cortical timing model of apparent motion than has so far been given; and (3) to ground a coherent model in an oscillatory framework, a general neurodynamical framework that has ever-expanding experimental support (Buzsaki, 2006, Canolty and Knight, 2010, Voytek and Knight, 2015).

The plan for this paper is to work from the details to the big picture to integrate the science and philosophy of immanent time consciousness (i.e., the specious present). After examining a differential latency account of apparent motion, thus squaring a key temporal illusion with a philosophically extensionalist position, I respond to specific objections. Then turning to a general framework, I describe the ROM neurodynamical model and show how it can account for the features of the specious present in a manner consonant with extensionalism. The final sections of the paper are devoted to canvasing a range of general philosophical objections.