Distrusting the present

Abstract

We use the hierarchical nature of Bayesian perceptual inference to explain a fundamental aspect of the temporality of experience, namely the phenomenology of temporal flow. The explanation says that the sense of temporal flow in conscious perception stems from probabilistic inference that the present cannot be trusted. The account begins by describing hierarchical inference under the notion of prediction error minimization, and exemplifies distrust of the present within bistable visual perception and action initiation. Distrust of the present is then discussed in relation to previous research on temporal phenomenology. Finally, we discuss how there may be individual differences in the experience of temporal flow, in particular along the autism spectrum. The resulting view is that the sense of temporal flow in conscious perception results from an internal, inferential process.

Bayes, time perception, and relativity: The central role of hopelessness

“…In order to reduce uncertainty and overall error, a hopeless/depressed mindset may recruit a strategy of central tendency that lengthens the judgement of shorter intervals and shortens the judgement of longer intervals (i.e., a Vierordt effect). This error reduction strategy is consistent with the general framework of Bayesian PEM and 'distrusting the present' Running head: HOPELESSNESS, TIME PERCEPTION AND RELATIVITY 27 (Hohwy et al, 2016). The present authors have looked at this process in more detail in another paper (Kent, Klein, & Van Doorn, in preparation).…”

“…In summary, evidence has been put forward in support of an association between symptoms of depression, time perception, and Bayesian PEM using a framework called "distrusting the present" (Hohwy et al, 2016). It was shown how key variables, namely autonomic arousal and hopelessness, have a combined effect on how individuals judged and experience time.…”

“…hopelessness. Here, 'distrusting the future' (h1) slows the process of 'distrusting the present' (h0) (Hohwy et al, 2016).…”

“…A comprehensive model of time judgement must also include 'memory' and 'decision' stages. One of the main aims of the present paper is to explain the apparent mismatch in depression between time experience and time judgement in terms of a Bayesian model of time perception, focused on the notion of 'distrusting the present' (Hohwy, Paton, & Palmer, 2016).…”

Duration Perception Versus Perception Duration: A Proposed Model for the Consciously Experienced Moment.

“…The model proposed here to account for duration perception therefore contrasts: 1) the first timescale up to 200 ms of the salience network that synchronises perceptual data into functional moments (Craig, 2009a): 2) the second timescale between 200 -1250 ms of the The trajectory of informational flow within Figure 1 is then illustrated in Figure 2 and elaborated as follows: 1) multimodal exteroceptive sensory data is received from asynchronous sources; 2) the internal interoceptive, autonomic, and cardiovascular rhythms of the anterior insular (Craig, 2009a(Craig, , 2009b and cortico-striatal-thalamic oscillators (Allman & Meck, 2012) of the salience network then synchronise incoming multimodal sensory Running head: DURATION PERCEPTION VS PERCEPTION DURATION 14 information within the first 125 -200 ms; 3) the homeostatic and motivational salience of the incoming perceptual data for the "sentient self" is then determined by the anterior insular cortex in the form of a "global emotional moment", here interpreted as global perceptual integration (p. 1934) (Craig, 2009a); 4) the anterior insular cortex then directs salient global perceptual information to either the DMN or central executive network depending on whether or not the information is required to perform a cognitive task (Goulden et al, 2014); 5) global perceptual information sent to the DMN undergoes automated processing of "rapidly selecting appropriate responses under predictable circumstances" (p. 12821) (Vatansever, Menon, & Stamatakis, 2017); 6) this automatic processing of learned patterns is a process of Bayesian inference that seeks to resolve incoming sensory-perceptual information (i.e., sensory likelihood distribution) with prior information (i.e., prior distribution) in order to make accurate predictions about the future with minimal error (Hohwy, 2013); 7) the process and rate of resolving sensory-perceptual information with prior experience leads to the experience of temporal flow via a Bayesian inferential mechanism called 'distrusting the present' (Hohwy et al, 2016); 8) the upshot of DMN Bayesian PEM is conscious perception (Hohwy, 2012); 9) on this account, consciousness is comprised of a relatively fast perceptual process (i.e., sensory likelihood) and a relatively slow thought-like process (i.e., prior distribution) (Koch et al, 2016); 9) salient information sent to the central executive network is attended to via working memory processes (i.e., episodic buffer) over intervals up to approximately 100 s (Goulden et al, 2014;Wittmann, 2011); 10) episodic memory contents are then integrated into declarative autobiographical memory and fed back into the DMN to regulate Bayesian PEM in the form of hierarchical prior distribution(s) and thought-like consciousness (Chen et al, 2013;…”