This week at The Junkyard, we're hosting a symposium on Jim Davies’ recent book:  Imagination: The Science of Your Mind's Greatest Power (Pegasus Books 2019).  See here for an introduction from Jim. See here for a commentary from Kourken Michaelian along with a reply from Jim. Today is a commentary from Mike Stuart along with a reply from Jim.

* * *

Commentary from Mike Stuart.

I’m very happy to be taking part in this symposium on Jim Davies’s book, Imagination: The Science of Your Mind’s Greatest Power. The book is aimed at scientists but could be read by almost anyone. It summarizes and interprets results of scientific studies that are relevant to imagination, as well as related topics including perception, memory, foresight, emotion, morality, hallucination, (day-)dreaming, visualization, imaginary friends, AI, and scientific creativity.

The subtitle of the book might bring to mind self-help books. And indeed, Davies provides concrete recommendations about how to use imagination to avoid storing painful memories (43), make better predictions about the future (60-75[1]), be a more effective altruist (75-6), be a better person (82-4), be happier (84-9), have lucid dreams (138-9), get better at sports (167), diet (168), remember (168-75), save your money (175-6), get healthy (177-81), and be more creative (188-90). Unlike most self-help books, however, Davies is often pessimistic (read: truthful). For instance, it seems that the power of our visual imagination begins to decline in our twenties. Davies summarizes a number of studies and methods for recovering or improving the visual imagination, but concludes that there is no evidence that any of them work (245-7).

As a philosopher, it’s probably best if I leave the science to the scientists. Instead, I’ll briefly discuss a number of Davies’s claims, asking how they might bear on philosophical issues.

Kinds of Imagination

Davies characterizes imagination as the creation of: “ideas in your head” (1), “possible situations,” (119), “mental constructions” (253) or “representation[s] of the world” (267), where these are composed (at least partially) from ideas, beliefs and memories, as opposed to external stimuli (2). This is quite inclusive. For example, any idea that comes to mind, voluntarily or otherwise, looks like it will count as an imagining, and Davies indeed allows that many perceptions (and all memories) are imaginings. At one point this leads him to wonder if “putting a clear boundary between what’s imagination and what’s not is impossible” (37). Such an encompassing characterization makes imagination almost synonymous with thinking, and erases useful distinctions between imagining and related states like supposing, conceiving, and entertaining.[2]

Getting more specific, Davies splits imagination into sensory and “conceptual” (i.e., non-sensory). Davies focuses on sensory imagination (mental imagery) for most of the book, which is the “clearest and most obvious” form of imagination (2). Sensory imagination is the generation of mental images (or mental sounds, feels, tastes, etc.), where those images are not caused by what they would be caused by in “real” perception. On this characterization (again, perhaps controversially), hallucinations, afterimages and dreams can be treated as imaginings.[3]

Judging by this book, cognitive scientists seem to mostly focus on (what philosophers call) “objectual” sensory imagination. Objectual imagination concentrates on objects or actions, and philosophers distinguish it from propositional imagination, which focuses on states of affairs. Thus, we objectually imagine a jar of peanut butter, and propositionally imagine that a jar of peanut butter conceals a spring loaded gag snake. Meanwhile, philosophers tend to focus on propositional imagination. Why is there this difference? It’s worth thinking about, especially as there is no philosophical consensus on how objectual sensory imagination relates to propositional imagination. Objectual sensory imagination is also epistemologically relevant. In my own research, I’ve found that this kind of imagination plays a large role in scientific progress (especially in biology), and this needs explaining. For those philosophers interested in objectual sensory imagination, this book collects just about everything contemporary science can tell us about it, laying the groundwork for the magic of interdisciplinarity.

Visual vs. Spatial Imagery

One important distinction that Davies highlights is between the visual and spatial aspects of mental imagery. “Visual information is about how things look, in terms of shape, color, texture, etc. Spatial information is about where things are in relation to each other” (236). These different aspects of mental imagery (and perception) are handled by different neural systems, and studies show that one can be impaired while the other remains. Thus we can have a sense for where things are without having a mental picture of them (e.g., “Think about kicking off your shoes under the dinner table. At the end of the meal, you might feel around for them and put them back on, all without looking…Spatial knowledge can come from any of the senses and can also come from understanding texts, as when someone verbally describes to you the layout of their home” (238)). I think this distinction is important for those of us interested in the epistemology of imagination, especially since, as Davies points out, we handle some visual tasks better than spatial tasks. And it interacts with the philosophical literature on mental imagery at least in the following way. Marghertia Arcangeli (2019) distinguishes between mental imagery as a mode of imagining, and as the content of an imagining. The distinction between visual and spatial imagery gives us another dimension for the mode of mental imagining, since that mode can be imagistic in either a visual or spatial way.

Quarantine

One feature of imagination mentioned by many philosophers is that it is “quarantined” from belief and action (see, e.g., Gendler 2003, Leslie 1987, Nichols 2004, Nichols and Stich 2000, Perner 1991). That is, “imagined episodes are taken to have effects only within the relevant imaginative context” (Salis and Frigg forthcoming, 11). Thus, imagining that someone lied to you does not make you believe that they really lied to you, and you don’t act suspiciously towards that person. And yet…

As Davies points out, imaginings do not come quarantined. Rather, quarantining is a process, and it’s not a simple one. We constantly monitor our mental activity for signs that something should be quarantined, and meanwhile, many of our imaginings do, in fact, attempt to activate action processes (i.e., break out of quarantine). It’s just that those processes are inhibited by other systems before the action is actually carried out.

Davies also notes that we don’t just quarantine imagination from belief: we also quarantine some imaginary scenarios from others, and such quarantining will often be partial, or nested. For example, we quarantine our imaginings about zombies from daily life, but we also quarantine different kinds of zombie imaginings from each other (e.g., where the cause of zombiehood is supernatural or a disease, whether the zombies are intelligent or not, etc.). When we encounter a new zombie fiction, it takes time to see what it should be quarantined from and what it should not.

Davies takes this to be evidence against the “pretense box” account of imagination developed by Nichols (2004) and Nichols and Stich (2000), according to which we have a box containing our imagined propositions (the pretense box), whose contents are quarantined from those of the “belief box,” in a way that allows a single set of inference mechanisms to govern both boxes.

Davies claims that there must be evolving sub-boxes within the pretense box, and the walls of the box itself must be flexible. I think this is an important complication. Though for what it’s worth, I’m not sure Nichols or Stich, or others, would disagree.

Limits of Imagination

In the final chapter, Davies asks “what kinds of things can we imagine, and what is impossible to imagine?” (256). He begins with visual imagery, and claims that “some things are just too abstract to create an image of” (256). Thus, we can visually imagine a chair, but not furniture; a judge, but not justice. This is tricky, because I think there’s a sense in which an imagining of a chair is an imagining of furniture, if that’s what it’s intended to be. In my imagination I make the chair “stand for” furniture, in a metaphorical way. Also, if this is a constraint on imagination, it’s not one that comes from the science of imagination, but philosophical considerations about what it is possible to represent via images and the limits of metaphor.

Other limits on visual imagination come from our previous experience. For example, “when you imagine a car accident, you don’t have to consciously put the cars on roads, they just seem to appear out of nowhere” (257). We also typically import all the usual laws of nature into our imagined scenarios (258). While this is generally true, I don’t see it as a limit on what we can imagine. If we want to, we can imagine the roads on top of the cars, or pretty much anything else. If we couldn’t, artists and scientists would never be able to imagine the world being any other way than they have experienced it to be in the past, which thankfully they can.

Perhaps Davies only means that the unconscious, involuntary elements of imagined scenarios tend to be constrained by previous experience. This idea goes nicely with discussions in the epistemology of imagination about “architectural” constraints on imagination, which are those constraints we inherit from evolution and our neural make-up (see Kind and Kung 2016, 21; Langland-Hassan 2016; Van Leeuwen 2016), as well as my own suggestion that perhaps we need one epistemology for conscious and voluntary imagination, and another for unconscious involuntary imagination (Stuart forthcoming).

Stray thoughts and observations

• Davies considers synesthesia to be a form of hallucination (120-1), but he doesn’t say what role, if any, imagination plays in such hallucinations. I’m curious to know what he thinks about this.

• Davies mentions a case of an aphantasic who can get sexually aroused using his imagination, only if he tries (and inevitably fails) to imagine arousing imagery. Davies hypothesizes that this person does have mental imagery, but no access to it. Rather than argue that this explains aphantasia in general, he claims that there might be two kinds of aphantasics: those who don’t have mental images (and can’t experience them), and those who do have mental images (and can’t experience them) (241-2). It’s unclear how this claim relates to the existing data (e.g., Fulford et al 2018), but if it’s true, this would be an interesting addition to the growing literature on aphantasia.

• The book offers a lot of evidence for accounts of imagination that take embodiment seriously. This is a good thing, as embodied accounts of cognition continue to gain in popularity elsewhere in philosophy of mind and epistemology.

* * *

Thanks to Julia Langkau and Kiran Phull for comments on an earlier draft of this post.

[1] For example, “The lesson is to imagine what you should be doing to achieve your goals, rather than fantasizing about them being achieved” (71).

[2] To be fair, I can’t come up with a good definition either. My best attempt stops short at a schema that provides different definitions of imagination, depending on how you fill out its parameters (Stuart forthcoming).

[3] One last thing to note about Davies’s definition of imagination and imagery is that they both focus on the generation of ideas and images. But I would think that imagination doesn’t just generate such things, imagining is also what we do when we explore and manipulate them in our minds.

References:

Arcangeli, M. 2019. “The Two Faces of Mental Imagery.” Philosophy and Phenomenological Research DOI: 10.1111/phpr.12589.

Bokulich. A. 2001. “Rethinking Thought Experiments.” Perspectives on Science 9: 285-307.

Fulford, J. Milton, F., Salas, D., Smith, A., Simler, A., Winlove, C., and Zeman, A. 2018. “The neural correlates of visual imagery vividness – An fMRI study and literature review.” Cortex 105: 26–40. DOI: 10.1016/j.cortex.2017.09.014.

Gendler, T. S. 2003. “On the relation between pretense and belief.” In Imagination, Philosophy, and the Arts, eds. Matthew Kieran and Dominic MacIver Lopes. London: Routledge.

Kind, A. and Kung, P. 2016. “Introduction.” In Knowledge through Imagination, ed. Amy Kind and Peter Kung. Oxford: Oxford University Press.

Langland-Hassan, P. 2016. “On Choosing What to Imagine.” In Knowledge through Imagination, ed. Amy Kind and Peter Kung. Oxford: Oxford University Press.

Nichols, S. 2004. “Imagining and Believing: The Promise of a Single Code.” Journal of Aesthetics and Art Criticism, 62 (2), pp. 129–39.

Nichols, S. and Stich, S. 2000. “A Cognitive Theory of Pretense.” Cognition 74: 115-47.

Perner, J. 1991. Understanding the Representational Mind. Cambridge: MIT Press.

Sorensen, R. 1992. “Thought Experiments and the Epistemology of Laws.” Canadian Journal of Philosophy 22: 15–44.

Salis, Fiora and Roman Frigg. Forthcoming. “Capturing the Scientific Imagination.” In The Scientific Imagination: Philosophical and Psychological Perspectives, ed. Peter Godfrey-Smith and Arnon Levy. Oxford: Oxford University Press.

Stuart, M. T. forthcoming. “Towards a Dual Process Epistemology of Imagination.” Synthese DOI: 10.1007/s11229-019-02116-w.

Van Leeuwen, N. 2016. “The Imaginative Agent.” In Knowledge through Imagination, ed. Amy Kind and Peter Kung. Oxford: Oxford University Press.

Response from Jim Davies: Limits of Imagination

Stuart's insightful overview of my book brings up the issue of the limits of imagination. I argue (256) that one cannot create a mental image of an abstract category like furniture; when one tries they must resort to a concrete image, such as that of a chair. Stuart brings up the fact that for an individual imaginer, that chair can stand for furniture, through some symbolic link or metaphor. 

I agree. For environmentalists, polar bears represent climate change problems, and for many Americans Richard Nixon represents shady dealings. What I was trying to communicate is that if you create a chair and attach the meaning of furniture to it, that connection is a further psychological step. I know this is true from my own experience. In my discussion of memory palaces (172), I describe the difficulty of creating images to represent abstract concepts: if I want to remember to listen to an MP3 that Liz sent me, I cannot image an MP3 file, so I picture Liz dancing with an iPod. I attach the meaning of the MP3 to the iPod when I encode the memory, but at recall time I forget this connection, and wonder why I wanted to remember Liz dancing with an iPod. But I never recall the image of an iPod and wonder whether or not it’s an iPod. I can recognize concrete things from the mental image without having to do a separate recall of its intended meaning. Perhaps I would have better described this limit as a limit on imagery, not on imagination in general. 

Another limit on imagination I brought up was that we can only imagine things and recombinations of things we’ve experienced before. Stuart counterpoints that we can imagine roads on top of cars with no trouble. My response is that this is still a recombination of previous experience in the sense that we’ve seen cars, roads, and experienced the spatial relationship of one thing being on top of another. So even if we have never seen a road on a car, we can combine these three concepts and imagine it. This is possibly an unfalsifiable claim I’m making here, because I could, post-hoc, describe any conceivable imagining in terms of abstract properties, events, objects, and relationships that someone has seen before. 

However, I have experienced myself how imagination can blossom after having an experience of something new. When I was a child, I heard about the game Dungeons & Dragons but had not access to it. I tried to create what I thought it would be and made a very simple board game where you roll dice to move around. When I actually played D&D my mind was blown. It was a kind of game I felt sure I would have never come up with given the pre-D&D experiences I’d had. After I understood what role-playing games were like, I started designing them myself. Now, Gygax and Arneson, who invented D&D, were able to come up with the idea, but presumably they’d been drawing on previous experience too--with wargames and other ideas. 

In part my argument is definitional--if imagination is generated from memories, then they must be constrained by the memory in some way. And populating memory with new ideas provides more fuel for the imagination engine. 

References

Hunter, E. C., Sierra, M., & David, A. S. (2004). The epidemiology of depersonalisation and derealisation. Social psychiatry and psychiatric epidemiology, 39(1), 9-18.