Phenomenal Consciousness Disembodied

Is an entity's physical constitution a central principle of folk psychology that guides judgments about phenomenal consciousness?  In a spooky new paper with Mark Phelan, we continue our examination of experiential state ascriptions by turning to the phantasmally disembodied—ghosts and spirits.

Lacking in any body whatsoever, spirits constitute the ultimate test of the basic embodiment view. If embodiment is a crucial cue for phenomenal state attribution, then we should expect important differences in ascription between human beings, on the one hand, and disembodied ghosts and spirits, on the other—just as we expect (given our prior work in this area) to find important differences in phenomenal state attribution for functional information (information about the goals, desires, etc, of an entity). However if functional information tends to cue mental state ascription independently of whether the entity has a physical body, then it undermines the embodiment hypothesis. This is what we set out to investigate, using spirits as our medium.

In five experiments, our results suggest that embodiment is not a central principle guiding ascriptions of phenomenal consciousness to these sorts of entities, while also continuing to support the important role of functional considerations in theory of mind judgments. We speculate that these findings may also at least begin to question the widespread nature of intuitions used to motivate absent qualia arguments against functionalism.

I notice there hasn't been much conversation on the blog lately, so some comments would really raise our spirits!

Comments

[John Turri]

You're right, the comment threads on this blog have been a ghost town lately ...

... but seriously, I love this paper. Great job, Mark and Wesley!

I had a further idea. You write, "If embodiment is a crucial cue for phenomenal state attribution, then we should expect important differences in ascription between human beings, on the one hand, and disembodied ghosts and spirits, on the other." Another way of testing this would be to run cases with two agents, one embodied and one disembodied, both of whom have the same sort of functional profile. Then have participants decide which option is better (or which is more likely, etc): that the embodied agent has attitude A, or that the disembodied agent has attitude A.

For example, perhaps you could devise a case about someone with one living uncle and one dead uncle. Have something happen to this person, and then ask which better describes the case (or something along those lines): that this event made the living (embodied) uncle happy, or that this even made the dead (disembodied) uncle happy.

If participants don't select the embodied uncle at significantly higher rates, then it supports the view that embodiment isn't a crucial cue to folk phenomenal-state ascription. But if participants do select the embodied uncle at higher rates, then it supports the view that embodiment is a crucial cue.

Anyway, that's not intended as a criticism of anything you've done in the paper. Rather, it's just a further thought about what is a fascinating and valuable series of studies.

[Mark Phelan]

Thanks for the comments John. Your suggestion is entirely revenant to our project. However, it's not entirely clear that higher rates of selection of the embodied uncle in the described scenario would support the embodiment view over the analytic functionalist account we favor. As we emphasize in our last paper, according to functionalism, physical embodiment may matter to ordinary attributions of particular mental states, simply because some sensory stimuli or behavioral outputs may be thought to require specific physical organs or apparatus (e.g., feeling hunger may be thought to require empty stomach inputs). In other words, information about body may itself supply functional information. What the proponent of the embodiment hypothesis needs is evidence that embodiment matters over and above functional information. This may be ascertainable using some method similar to that you have described. But in many cases it will be difficult to tell whether such results actually support embodiment and aren't simply another apparition.

[John Turri]

Hi, Mark. I agree that it's important to ensure that information about embodiment isn't simply providing (relevant) information about function. Otherwise, my suggestion would be tantamount to chasing phantoms.

Nevertheless -- and I hope I'm not being too cryptic -- it still seems that there is a specter of further hope for the embodiment theorist to potentially wring some supporting evidence out of this approach. Suppose that something wonderful happens to the character in the story. Make it so that Dead Uncle and Live Uncle are both aware of this. At that point, is there any reason to think that specifically functional information provided by embodiment makes Live Uncle seem like a better candidate for being happy than Dead Uncle?

[Wesley Buckwalter]

Thanks, John! We were looking for additional ways to test for literal ascriptions using direct comparisons, and your uncle cases sound really great. Setting Mark's worry about functional information aside, we would definitely want to predict that people would choose embodied and disembodied uncle at similar rates. Perhaps the forced choice question between the two might provide some weird pragmatic pressures or something, but if there was a 'boo-th' option, I would think we would see response rates higher than chance.