The Visualizer’s Fallacy

Most aphantasics will be familiar with the following situation: You are trying to explain the phenomenon of aphantasia to a non-aphantasic, and they react with a mix of confusion and curiosity. At best, your conversation partner will show genuine interest and view this as an opportunity to learn more about your experience. At worst, they will simply not believe you (we could refer to this as aphantasia skepticism). But in either of the two scenarios (most cases likely fall somewhere in between), one question might pop up: “If you do not have mental imagery, how can you even think?” 

In this article, I will argue that this notorious question, as well as a range of other related questions of the form “If you are aphantasic, how can you…?”  indicates a wide-spread bias held by many (if not most) visualizers, which I will refer to as the visualizer’s fallacy (Scholz, 2023). 

The visualizer’s fallacy describes the idea that because visualizers experience visual mental imagery when carrying out a certain task (e.g., thinking, drawing, etc.), they falsely assume that aphantasics, due to their inability to form visual mental imagery, cannot solve said task. In the remainder of this article, I will illuminate the historical roots of this fallacy, show which (implicit) assumptions its logic is built on and, lastly, explain how the fallacy can be attacked. 

The Origins of the Visualizer’s Fallacy

In 2015, Adam Zeman famously coined the term aphantasia, which he defined as “a condition of reduced or absent voluntary imagery” (Zeman et al., 2015, p. 379). In the same article, he explains that the term is inspired by the classical Greek term “phantasia”, which the Ancient Greek philosopher Aristotle used to describe the faculty of imagination. Zeman then goes on to say that he prefixed Aristotle’s term “phantasia” with an “a”, in order to denote the absence of (visual) mental imagery and thus “aphantasia” was created. 

The link between Aristotle and aphantasia is intriguing, because Aristotle, in ca 350 BC, wrote that “the soul never thinks without an image”(Aristotle, De Anima III, §431a), thus representing one of the first accounts that took mental images to be a necessary condition for thought.¹ So, we see that the very term aphantasia is historically linked to the idea that thinking necessarily requires mental images. And this idea, in turn, would imply that aphantasics, who do not have mental images, are incapable of thought.²

Of course, just because Zeman was inspired by Aristotle’s term, that does not mean that he agrees with Aristotle’s ideas. However, Zeman himself has repeatedly expressed his surprise that aphantasics function well on a range of cognitive tasks. In 2020, for example, he wrote that “[t]he discovery that some people get along fine in the absence of visualization […] is striking” (Zeman, 2020, p. 706). Thus, while he probably did not think that aphantasics are incapable of thought in general, he at least expected them to be impaired with respect to the  performance of a number of cognitive tasks.

The Mental Rotation Task – Striking Findings

One of the tasks which aphantasics were not expected to be able to solve is the so-called mental rotation task, where participants are presented with drawings of a number of three-dimensional figures and are then asked whether the depicted figures are the same figure, only viewed from different angles of rotation. (See Figure 1).

Since mental rotation tasks were widely understood to test for visual mental imagery (Richardson, 1999), one might assume that aphantasics could not solve this task. However, by now there are multiple studies that show that aphantasics can solve mental rotation tasks (Pounder et al., 2022; Crowder, 2018; Zeman et al., 2010). Even more so, in all of these studies, aphantasics were found to give correct answers more often than visualisers, although they were also shown to take longer to solve the task.

There are also other tasks which were assumed to rely on mental imagery and which aphantasics were found to be able to solve. For example, Zeman and his colleagues asked a patient with acquired aphantasia, “Is the green of grass darker than the green of a pine tree?”(Zeman et al., 2010, p. 147).³ While it might seem strange to an aphantasic that such a question is expected to require the use of mental imagery, the common assumption seems to have been that this question can only be answered by means of forming mental images of grass and pine trees and then reading the correct answer off the mental image, so to speak. But where did this assumption come from?    

The Logic Behind the Visualizer’s Fallacy 

When a visualizer is solving the mental rotation task, they report that they visualize the first figure and then rotate that figure ‘in their mind’, in order to see whether it matches with any of the other figures. Similarly, when a visualizer answers the question about the different shades of green of grass and pine trees, they report visualizing grass and pine trees and then comparing the colors they see in their mental images. Thus, when visualizers solve these kinds of tasks, they are having a so-called quasi-perceptual experience, meaning an experience as if they were seeing something, only with their ‘mind’s eye’ (the rotating figure or the grass and pine trees, respectively).  

In order to understand the logic behind the visualizer’s fallacy, we must now understand two assumptions about this quasi-perceptual experience. Firstly, visualizers assume that they solve the task in question (for example the mental rotation task) because of this quasi-perceptual experience. So, they think that the mental image is causally related to their own performance. Secondly, visualizers assume that there are no other strategies for solving the task (that is, no strategies that do not involve the quasi-perceptual experience). With these two assumptions in place, together with the third assumption that aphantasics cannot visualize, the visualizer’s fallacy, with respect to a specific task, is derived as follows:  

Assumption 1 (A1): The quasi-perceptual experience is essential for the visualizer’s strategy.

Assumption 2 (A2):  The visualizer’s strategy is the only viable strategy for the task.

Assumption 3 (A3):  Aphantasics cannot visualize.

Conclusion 1 (C1):  Aphantasics cannot use the visualizer’s strategy. (follows from A1 and A3)

Conclusion 2 (C2):  Aphantasics cannot solve the task (follows from C1 and A2)

Now that we have understood the implicit assumptions underlying the visualizer’s fallacy, we can attack and reject it. 

2 Ways of Rejecting the Visualizer’s Fallacy

As we have seen, the visualizer’s fallacy consists of three assumptions. Assumption 3 states that aphantasics cannot visualize and this is true (by definition). However, the two other assumptions can be attacked, and thus there are two ways of rejecting the fallacy.

The Weak Rejection

One way to reject the fallacy is to attack the assumption that the visualizer’s strategy is the only viable strategy to solve the task (A2). To attack this, we can propose that aphantasics are simply using alternative strategies, in order to solve the task. For example, aphantasics might make use of their spatial sense, in order to solve mental rotation tasks, and they might make use of semantic memory when answering the question about the different shades of green. If we allow that there are strategies that do not rely on visualization, we reject assumption 2 and thus conclusion 2 does not follow anymore, since there is no reason anymore to assume that aphantasics cannot solve the task.

The Strong Rejection

Another way to reject the fallacy is to attack the assumption that the visualizer’s strategy essentially involves the quasi-perceptual experience (A1). Here, instead of saying that there are alternative strategies, we argue that even for the visualizer, the quasi-perceptual experience is only an accompanying phenomenon. This means that while the visualizer might have the experience as if she were seeing the figure rotating, she is not solving the task because of this experience. In the case of mental rotation tasks, for example, we might suggest that aphantasics and visualizers both use their spatial sense in order to solve the task and that the only difference is that the visualizer then also has a quasi-perceptual experience.⁴

Similarly, in the case of the question about the shades of green, one might suggest that instead of reading the correct answer off the mental image, the images of the shades of green can only be formed if one already knows how grass and pine trees look like (for a more in-depth analysis of this approach, see Scholz, 2023, section 4.2.1). So, if we can show that aphantasics and visualizers essentially use the same underlying strategy to solve the task, we can reject the conclusion that aphantasics cannot use the visualizer’s strategy (C1) and then the conclusion that aphantasics cannot solve the task (C2) would not follow.

Importantly, both the weak and the strong rejection strategies are sufficient in order to reject the visualizer’s fallacy with respect to a specific task. Can aphantasics solve the mental rotation tasks because they use an alternative strategy, or is it because not even the visualizer really needs their mental image? At this point, we do not know. But what we know is that aphantasics can solve mental rotation tasks, showing us that visualizers were wrong to assume that they cannot (and the same, obviously, goes for thinking in general as well).  

Concluding Remarks 

We are all heavily biased due to our own experience of the world, and imagination is a prime example of this. Visualizers have the experience that they solve a wide range of cognitive tasks by means of visualisation. Even more so, since their everyday thinking is often accompanied by mental images, they assume that thinking, for all people, necessarily relies on mental images. However, as the past decade of research into aphantasia impressively shows, people vary in their individual experience and for aphantasics, imagination, as well as thinking, occurs in the absence of mental images, thus showing us that mental images are not as important as we might have thought. 

Yet, since aphantasia is still a very recent and thus not yet well-known phenomenon, and because most people are visualizers, it is hard for them to shake off their own belief regarding the high importance of mental images for our cognition.

And so, the next time you meet someone who asks you the notorious question of how you, as an aphantasic, can even think (or do any other cognitive task), just ask them why they are so surprised by this. You will see that their surprise is grounded in their own experience, and you will be able to identify the underlying logic of the visualizer’s fallacy in their thinking. And then, by means of questioning the assumptions of the fallacy, you will be able to give your conversation partner an opportunity to not just learn about the variety of personal experience but also to rethink some of their own biases with respect to the importance of mental images.

Footnotes

1. The classical Greek term which Aristotle used and which is nowadays translated as ‘image’ is ‘phantasma’ [pl. Phantasmata]. The original definition which Aristotle gives for the term ‘phantasia’, which is nowadays translated as ‘imagination’ is “that in virtue of which we say an image [phantasma] arises in us” (De Anima III, 428a 1-2). ↩︎
2. In this article, I am using the term ‘mental image’ synonymously with a quasi-perceptual experience (that is, an experience as if seeing), or with what is sometimes referred to as ‘seeing with the mind’s eye’. And while most psychologists and philosophers indeed view this quasi-perceptual experience as a necessary part of mental images, I have argued elsewhere (see Scholz, 2023, section 4.2.3) that mental images can also be understood in a more pragmatic sense that does not have to involve any sort of quasi-perceptual experience. ↩︎
3. The taxonomy of aphantasia distinguishes between life-long and acquired aphantasia. Persons with life-long aphantasia (also referred to as congenital aphantasia) do not recall ever having had mental imagery. This group represents the majority of people with aphantasia and is estimated to consist of 2-5% of the overall population.  Persons with acquired aphantasia, on the other hand, had mental imagery but lost it later in life, for example, after undergoing brain damage (this is referred to as neurogenic aphantasia) or a mental disorder (this is referred to as psychogenic aphantasia). For a discussion of the neurological basis of both lifelong and acquired aphantasia, see Scholz (2023), section 3.2.3, or Zeman (2020). ↩︎
4. In the case of the mental rotation task, there is indeed some experimental evidence suggesting that visualisers are only really making use of spatial information when solving the task (see Liesefeld & Zimmer, 2013). ↩︎

References

Aristotle. (c. 350 BC/1984). Complete works of aristotle, volume 1: The revised oxford translation (Vol. 1). Princeton University Press.

Crowder, A. (2018). Differences in spatial visualization ability and vividness of spatial imagery between people with and without aphantasia (Doctoral dissertation). Virginia Commonwealth University.

Liesefeld, H. R., & Zimmer, H. D. (2013). Think spatial: The representation in mental rotation is nonvisual. Journal of experimental psychology: Learning, memory, and cognition,39(1), 167–182.

Pounder, Z., Jacob, J., Evans, S., Loveday, C., Eardley, A. F., & Silvanto, J.(2022). Only minimal differences between individuals with congenital aphantasia and those with typical imagery on neuropsychological tasks that involve imagery .Cortex, 148, 180–192.

Richardson, J. T. E. (1999). Imagery. Psychology Press.

Scholz, C. O. (2023). Imaginability as Representability: A Wittgensteinian Approach to Aphantasia. Master of Logic Thesis (Mol) Series.

Zeman, A. (2020). Aphantasia. In A. Abraham (Ed.), The Cambridge handbook of the imagination (pp. 692–710). Cambridge University Press.

Zeman, A., Della Sala, S., Torrens, L. A., Gountouna, V.-E., McGonigle, D. J., & Logie, R. H. (2010). Loss of imagery phenomenology with intact visuo-spatial task performance: A case of ‘blind imagination’. Neuropsychologia,48(1), 145–155.