Research Summary: The Subjective Experience of Time

post by Jason Schukraft · 2020-08-10T13:28:13.181Z · EA · GW · 6 comments


  Main Text


Considering differences in the subjective experience of time may affect the proportion of resources we wish to allocate to different species. I’ve recently written two pieces on the subject, here [EA · GW] and here [EA · GW]. Counting somewhat conservatively, the two posts total more than 23,000 words. In this (much shorter) post I attempt to succinctly introduce the topic and convey my main conclusions. The aim is to create a digestible overview of my current thinking that imposes a smaller time burden on readers.

Main Text

If you’ve ever had the misfortune of being in a car accident or fighting in a war zone or being attacked by a wild animal, you may already be familiar with putative differences in the subjective experience of time. When confronted with life-threatening circumstances, humans often report that time seems to slow down. Events that are over in tens of seconds seem to stretch on for minutes, allowing rapid assessment of the scene and quickfire decisions that, in some cases, save one’s life. These types of differences are also sometimes induced artificially. An LSD trip might seem to extend for days when in fact it was over in an afternoon. If these sorts of events constitute genuine differences in one’s temporal phenomenology, they change the quantity of subjective experience one undergoes in a given period of objective time.

There is intriguing empirical evidence that individuals of different species differ in their experience of time. So, for example, a zebra finch (a type of small bird) might characteristically experience time the same way a human does during the moment of impact of a car accident: a whole lifetime played out in (what humans would describe as) slow motion. On the other hand, some animals might experience the opposite: if we could adopt the temporal viewpoint of, say, a leatherback turtle, the world might appear as if it were sped up 4x.

Such differences—if they exist—would affect the moral value of experiences. For a given negative stimulus of the same subjective intensity and objective duration, animals with a faster rate of subjective experience, like the zebra finch, would suffer more than animals with a slower rate of subjective experience, like the leatherback turtle. The zebra finch would experience the painful stimulus as longer, in a robust sense, than the leatherback turtle. That is, over the same objective duration, the finch would experience a greater number of painful moments than the turtle.

Again, we don’t know for certain which animals—if any—experience time at different rates, and because such potential differences concern private, subjective experience, we can’t ever be certain such differences exist. Nevertheless, there are tantalizing clues which may enable us to draw tentative conclusions.

Some of the clues are behavioral:[1] ornithologists have long studied the temporal complexity of birdsong, with its extremely short fundamental periods (measured in mere milliseconds), much shorter than what humans can produce or identify. In fact, avian ethologists often claim that humans cannot truly appreciate birdsong unless the tape is slowed down.

Some of the clues are neurological. If the brain is like a computer, we might be able to measure the brain’s clock speed by looking at the speed at which it can transmit signals. Parrot and songbird brains pack many more neurons per cubic millimeter than elephant and whale brains, allowing the smaller, denser bird brains to transmit signals much faster.

Some of the clues concern differences in temporal resolution, which is the rate at which a perceptual system samples information about its environment. Imagine a light alternating between on and off; if the alternation is fast enough, you’ll cease to see a flickering and instead see a steady glow. The threshold at which this change occurs is called your critical flicker-fusion frequency (CFF), and it is a measure of visual temporal resolution. CFF thresholds have been determined for over 70 species. Humans typically perceive the switch from flickering to steady glow at around 60 Hz. For some songbirds, the comparable figure is 146 Hz.

We may soon even be able to identify rough neural correlates for the subjective experience of time. In humans, differences in neural oscillations in the alpha-band (brainwaves in the 8-12 Hz range) are correlated with differences in reaction times, CFF, and temporal integration windows. I talked to one neuroscientist who wants to extend this work to see if natural variation in alpha-band oscillations predicts differences in duration estimates. He thinks that with the right funding, the experimental paradigm could eventually be extended to investigate temporal experience in many nonhuman animals.

Considering differences in the subjective experience of time could change the way we allocate resources to different types of animals. Currently, if one wants to approximate differences in moral status or capacity for welfare using a measurable metric, the best one can do is appeal to neuron count, brain-to-body-mass ratio, encephalization quotient, or something similar. One interesting aspect of the temporal experience proxies I’ve identified is that the rankings of animals they generate differ fairly dramatically from rankings based on brain-size metrics. Honey bees only have about a million neurons, but their rate of subjective experience appears to be much faster than big-brained animals like elephants. Incorporating consideration of differences in the subjective experience of time into our prioritization process (perhaps weighting temporal experience metrics equally with brain-size metrics) would change our overall ranking of animals.

By reading the two posts in full, you’ll learn why I believe:

This area of research is admittedly quite speculative. I spoke to several experts who suggested that making concrete progress on this topic, while not impossible, would be difficult. On the other hand, this area of research is also quite neglected. Although there is a rich literature in adjacent subjects, there have only been a handful of scientific studies that attempt to directly probe differences in the subjective experience of time in humans, and none attempted to extend the investigation to nonhuman animals. By bringing together disparate threads from related fields, we may be able to encourage interdisciplinary research that begins to unravel the mystery of temporal experience.


This essay is a project of Rethink Priorities. It was written by Jason Schukraft. Thanks to Kim Cuddington, Marcus A. Davis, Derek Foster, Peter Hurford, Jeff Sebo, Saulius Šimčikas, and Daniela Waldhorn for helpful feedback on earlier drafts. If you like our work, please consider subscribing to our newsletter. You can see all our work to date here.


  1. These examples focus on birds because the relevant data already exist, but future scientific research can uncover similar clues for other kinds of animals. ↩︎


Comments sorted by top scores.

comment by ShayBenMoshe (shaybenmoshe) · 2020-08-10T19:55:06.558Z · EA(p) · GW(p)

Thank you for writing this summary (and conducting this research project)!

I have a question. I am not sure what the standard terminology is, but there are (at least) two different kinds of mental processes: reflexes/automatic response and thoughts or experiences which span longer times. I am not certain which are more related to capacity for welfare, but I guess it is the latter. Additionally I imagine that the experience of time is more relevant for the former. This suggests that maybe the two are not really correlated. Have you thought about this? Is my view of the situation flawed?

Thanks again!

Replies from: Jason Schukraft
comment by Jason Schukraft · 2020-08-11T01:27:24.589Z · EA(p) · GW(p)

Thanks, that’s a great question!

Welfare is constituted by those things that are non-instrumentally good or bad for the creature. Insofar as reflexes are unconscious, they probably are not non-instrumentally good or bad. (They are, of course, often instrumentally good; they help the creature get other things that are good for it.) Conscious experiences, on the other hand, are usually non-instrumentally good or bad. Experiences with a positive valence are non-instrumentally good; experiences with a negative valence are non-instrumentally bad. (Experiences that are perfectly neural may not be non-instrumentally good or bad; experiences can also be instrumentally useful in a variety of ways.)

Differences in the subjective experience of time—assuming they exist—are relevant to welfare (both realized welfare and capacity for welfare) because they reflect differences in the amount of experience a creature undergoes per unit of objective time. I write about the moral importance of the subjective experience of time in this part [EA · GW] of the first post.

You’re right that there are other aspects of temporal perception that may not be directly relevant to welfare. We already know that there are differences in temporal resolution (roughly: the rate at which a perceptual system samples information about its environment) across species. Enhanced temporal resolution may, among other things, enable faster unconscious reflexes. Naturally, the speed of a creature’s reflexes will indirectly contribute to its welfare, but those unconscious reflexes won’t be part of what constitutes the creature’s welfare. Whether or not there is a correlation between temporal resolution and the subjective experience of time is an open question, one that I explore in depth in the second post [EA · GW].

Hope that clarifies things a bit for you, but if not, please ask a follow-up question!

Replies from: shaybenmoshe
comment by ShayBenMoshe (shaybenmoshe) · 2020-08-11T19:28:07.914Z · EA(p) · GW(p)

Thank you for the eloquent response, and for the pointers to the parts of your posts relevant to the matter.

I think I understand your position, and I will dig deeper into your previous posts to get a more complete picture of your view. Thanks once more!

comment by MichaelStJules · 2020-10-25T02:48:46.473Z · EA(p) · GW(p)

Antonia Shann also wrote a summary of one of your post for Faunalytics:

Replies from: Jason Schukraft
comment by Jason Schukraft · 2020-10-25T13:51:34.970Z · EA(p) · GW(p)

Cool, thanks Michael, I hadn't seen that. (And thanks to Antonia as well for writing the summary!)

comment by deinprasad ( · 2020-08-25T07:35:40.693Z · EA(p) · GW(p)

Thank you for conducting this research and putting it in simple language, so that any non SME can also understand what it is about and what is your point view about it. If at all i can contribute in anyways to your work, i would be glad to do so. All the best to you for your future work.