Hoffman's Conscious Realism:
A Critical Review

3. Interface Theory of Perception (ITP)

3.3 Perceiving Distance

Book cover: Mindware: An Introduction to the Philosophy of Cognitive Science by Andy Clark

Now, granted, there are many times when the items in our sensory field—our 'icons'—don't reflect reality as it is. But does this fact entail that our percepts are always non-veridical? Hoffman et al [2015a: 1497] point to 'obvious cases where our perceptions radically disagree with our careful measurements'. They point to how the 'sun, moon and stars, for instance, all look far away, but they all look about equally far away'. Citing the research of other neuroscientists, they add that '[e]ven at close distances our perceptions differ from our careful measurements'.

I'll return to the case of perceiving distant objects shortly. But is it really the case that our perception of distance up close is always non-veridical? Hoffman et al [2015a: 1497] point to research studies that purportedly show this. In fact, they seem to show the opposite; that in many perceptual circumstances, our perception of distance is veridical. For example, Kappers [1999: 1001] studying haptic perception in subjects makes only the weaker conclusion that 'haptic perception of, for example, distances or parallelity does not always conform to physical reality'. And where there is deviation from reality, Kappers found an algebraic relation between the two. As she concludes: 'The results clearly show that what subjects feel as parallel deviates systematically from what is actually physically parallel [Kappers 1999: 1009]'. In Hoffman et al's [2015a: 1484] mathematical simulations of the evolutionary process, such algebraic relations (either isomorphic or homomorphic) count as veridical within a realist strategy.

Koenderink et al's [2010: 1163] study of the visual illusion resulting in perspective distortion cited by Hoffman et al [2015a: 1497] also reveals an algebraic relation (viz.: 'large systematic deviations') between perception and reality. And again, Pont et al's [2011: 6] study of depth perception finds an algebraic relation in the subjects 'adjusted foreshortenings as a function of distance and size'.

Now, these researchers sought to demonstrate how the particular perceptions that are the subject of their study are not 'veridical'. However, note how their use of the term 'veridical' is used in the sense of there being an identity relation between the subject's perception and reality. In contrast, Hoffman et al [2015a: 1483] define all of the versions of realism that they seek to model, from that of the 'omniscient realist' to that of the 'critical realist', as acting on either an isomorphic or the less stringent homomorphic mapping between perceptions and reality. It is a surprise, then, that Hoffman et al [2015a: 1497] end their citation of these studies with a wholehearted assent to Koenderink's conclusion that the 'very notion of veridicality itself . . . is void' [Koenderink 2014: 5]. Perhaps it is not so surprising given that Koenderink is a Hoffman acolyte.

A further point worthy of note here is that the researchers cited examined perceptual distortions under unusual conditions (e.g., one eye blindfolded, objects hidden by a table). However, there are many circumstances in which such distortions are wholly absent. For example, when I lay out marbles evenly spaced at particular points along a ruler laid out on the floor, what reason is there for thinking that my perception of the space between each marble is non-veridical? Under these kinds of normal perceptual conditions, neuroscientists, in fact, have found in the mammalian brain neuronal grid cells whose firings map out just such evenly spaced locations in the environment. (See, for example, Balkenius and Gärdenfors 2016, McDermott 2020 and Moser et al 2008.) This linear mapping between perception and the environment is what we would expect if perception of local distance is veridical.

Let me now return to the case of perception of objects very far away, such as the moon and the stars. For these objects, Hoffman offers a standard evolutionary account for why we misperceive their relative distance. As he responds during an interview with Frohlich [2019]: 'So the idea will be that evolution has shaped us with a very simplified interface that's been shaped mostly to report the stuff that's going to keep us alive.' He explains further by contrasting the evolutionary advantage gained from perceiving close distances accurately:

Space and time are just a data structure. They're there to represent fitness payoffs. The distance from me to an apple, say, here like two meters away versus another apple, you know, 20 meters away—that distance is coding the percentage of my caloric resources that I currently have that would be required to be expended to get the resources in the apple at two meters versus 20 meters. In other words, distance is a calorie expenditure representing fitness cost. And so, it's no surprise that the stars look about as far away as the mountain, because they're both at infinity given my caloric resources.

[Frohlich 2019]

Book cover: Enlightenment Now: The Case for Reason, Science, Humanism, and Progress by Steven Pinker

Here, Hoffman has let the cat out of the bag. To explain why perception of very large distances is non-veridical, Hoffman illustrates his point by contrasting this case with a situation in which veridical perception of distance is necessary for survival. In the case of getting an apple, evolution selects for accurate perception of distance, otherwise we would be expending uneconomical amounts of calories.

Hoffman could claw back here and simply insist that all perception of distance is non-veridical. This would not be a good move as it would render his and his team's own explanation of their thesis incomprehensible. For example, I refer you to Hoffman et al's [2015a] explanation of their FBT Theorem. To help us understand their thesis, they construct a number of explanatory diagrams. By the time we get to the diagrams, Hoffman and his team have already told us that using the strict interface strategy that we all do, 'none of our perceptions reflect the structure of the world' [2015a: 1484]. Our perceptual mappings to the world are neither isomorphic nor more liberally homomorphic, as they illustrate in their Fig. 3 (reproduced below) [2015a: 1486]. So, they instruct us, for an object of perception that we perceive as green, the actual scalar quantity of that object could be around 30 or 70. We just don't know and there is no way of knowing.

Hoffman's diagrams for plotting payoff vs resource quantity for critical realist and interface strategies

Now, apply that revelation to how we should interpret Fig. 2 on the same page (reproduced above). Look at the 'Resource Quantity' axis in Hoffman et al's diagram. If our perception of distance preserves no structure of that to which we are looking, then when we look at the '50' resource quantity point that appears half way along the horizontal axis between the '0' and '100' anchors, that point could in actuality be mapped to either around 30 or 70. According to Hoffman et al's ITP, we just don't know. If, ex hypothesi, colours do not map isomorphically or homomorphically to actual resource quantities and our perception of distance is just as non-veridical as perception of resource quantity, then our perception of points along a line likewise leaves us clueless about actual distance along a line. This is not just a matter of poor precision in sensing distance along Hoffman et al's drawn line. If we are to believe their radical thesis, on their own terms, for each point on the line, we have no idea whether that point indicates a greater or lesser distance compared with the points to its left and to its right. This creates a problem not only for making sense of Hoffman et al's diagrams, but also for our understanding of any chart that uses axis.

Copyright © 2020, 2022

You will be interested in

Share This

  • twitter
  • facebook
  • linkedin
  • googleplus
  • gmail
  • delicious
  • reddit
  • digg
  • newsvine
  • posterous
  • friendfeed
  • googlebookmarks
  • yahoobookmarks
  • yahoobuzz
  • orkut
  • stumbleupon
  • diigo
  • mixx
  • technorati
  • netvibes
  • myspace
  • slashdot
  • blogger
  • tumblr
  • email
Short URL:
PDF Download Hoffman's Conscious Realism: A Critical Review

Download this essay