A report by Andrea Blomkvist
The relation between imagination and perception is a topic which has drawn the attention of philosophers, psychologists and neuroscientists for some time now. However, unlike some topics where interest wanes with time, interest in this theme has clearly intensified. This much was evident from our most recent installment of the Northern Imagination Forum, with the theme “Imagination and Perception”, which was held online on the 20th of January 2022 and which drew over 40 participants from across the world. Our speakers – Fiona Macpherson, Nadine Dijkstra, Ophelia Deroy, and Elisabeth Camp – delivered highly relevant and stimulating talks on various aspects of the theme, such as the phenomenology of visual imagery and visual perception, the neural basis of reality-monitoring, how to explain experiences of extraordinary perception, and how artworks in different media deploy imagination to express perspectives.
Starting us off was Fiona Macpherson with her timely talk about the phenomenology of visual imagery and visual perception, a topic ripe for discussion in philosophy and psychology at the moment. What, if anything, actually distinguishes the phenomenology of visual perception from that of visual imagery? Many have thought that there has to be some categorical difference between these, for example, Hume argued that visual perceptual states are inherently more vivid than visual imagery. From there, we have also had a bunch of different suggestions for how they could categorically differ, for example in their determinacy, or with respect to the sense of reality or presence. But Macpherson argued against all of these, claiming that none of these differences categorically mark an experience of visual perception from one of visual imagery. Take vividness for example. Here, the standard view is that seeing something is more vivid than imagining seeing it, so seeing an apple is more vivid than imagining an apple (maybe the real apple appears redder and brighter, whereas the imagined apple is somehow dimmer and duskier?). But this crucially depends on which experiences we are comparing. What if we are comparing imagining a carnival to seeing a misty loch? Here, it seems like the former is more vivid than the latter. It also seems wrong to say that our experience of visual imagery is anything like the images used on 5-point scales to illustrate vividness. These often depict an object, such as an apple, being more or less vivid, where the apple goes from bright red to grey and dim, to nothing. But surely the imagery we have when we imagine an apple (a bright red one) isn’t equivalent to seeing a dull and dusky one? These scales simply do not seem to capture our phenomenal experience. Macpherson gave similar arguments for why we cannot categorically tell visual perception from visual imagery by means of their determinacy or sense of reality/presence either. Instead, she put forward the suggestion that we should not look for a categorical distinction between visual imagery and visual perception, but rather appreciate their non-categorical nature.
On a similar note, Nadine Dijkstra also discussed visual perception vs. visual imagery, with the aim of unravelling the neural mechanisms that dissociate imagination from reality. Dijkstra suggested four different neural mechanisms which might do this work: sensory processing, cognitive control, predictability, and sensorimotor contingencies. However, she argued that none of these factors can non-ambiguously tell us whether a state is perceptual or imaginative. Let’s illustrate the case with sensory processing. Here, Dijkstra showed that the total amount of sensory activation in the visual cortex is lower for visual imagery compared to visual perception, since the visual cortex is not activated in the same way during these two experiences. During bottom-up driven visual perception, incoming stimuli are received by the middle layers of the primary visual cortex, and all layers of the primary visual cortex carry information about what is perceived. Notably, this layer is the one which carries the most excitatory neurons. However, visual imagery is top-down driven, and hence there is no such activation in the middle areas for visual imagery, entailing that the total amount of sensory activation is lower. Hence, in general, the strength of sensory activation is lower for visual imagery than for visual perception, and this looks like a prima facie good case of a sensory signal that the brain could use to monitor whether a state is real or imagined. But Dijkstra further argued that this is unlikely to work as perceptual and imagery areas are largely overlapping and use the same format of representation (meaning that we can decode content in visual imagery in the same way as we decode perceptual content). But importantly, some people have extremely vivid visual imagery (e.g. hyperphantasics), which is increasingly perception-like in its strength and format of signal, but they are still able to tell that they are imagining. Hence, sensory processing cannot do the reality-monitoring work after all. The upshot of Dijkstra’s talk was that none of the proposed neural mechanisms could perform reality-monitoring on their own, and instead, we need to appeal to a high-order metacognitive monitoring system. The idea is that this system evaluates first-order signals from all the other relevant neural areas, and different types of source confusions could arise from deficits in the different components of the system.
Also interested in reality, Ophelia Deroy focused on how we can explain extraordinary perception – perceptual experiences that we just cannot believe, like seeing a double rainbow. In cases of extraordinary perception, there is no doubt that these perceptual experiences are taken to be real by the subject, but nonetheless there is some kind of resistance to them. Deroy argued that these experiences are distinct phenomena – they are not just cases of hallucination or confusing imagination for reality, and she demonstrated many interesting cases of extraordinary perception. For example, these experiences can be generated in the lab using multisensory stimuli, such as the McGurk effect which arises when the syllables we hear are incongruent with the movement of the lips of the speakers. Other experiences include derealisation, where Type 2 derealisation is characterised by the subject believing that the object they perceive is real, but they still experience some friction with regards to this belief. So how can we explain the cases of extraordinary perception; why do these cases arise? Deroy argued that the best way to explain these experiences is offered by a composite account, where the sense of reality in perception combines a categorical tag (“this is perception”) with a combination of disparate feelings, including metacognitive feelings. Going back to the case with the double rainbow, she explained as follows: there is a categorical tag which categorises the experience as perception, but this is combined with e.g. a low metacognitive confidence in the visual experience as real.
Finally, Elisabeth Camp showed how the perception of artworks and interpretation of text in different media deploy imagination in different ways through the role which perspective plays in interpreting content and focusing attention. Camp opened by asking us to compare the passage from Genesis about Isaac’s sacrifice to the visual rendition of the same scene by Caravaggio. These two artefacts, although representing much of the same content, differ perspectivally, and they encourage a different interpretation of the content that they represent. Camp proceeded to define a notion of perspective that, whilst preserving key insights about how the concept is applied in perception, can also capture higher-level cognitive phenomena. According to her, a “perspective” is a disposition to interpret – a style of noticing, relating and responding to a certain part of the world. Whilst pictures elicit a perspective by means of their compositional structure, literary narratives elicit a perspective by selecting, describing, and connecting abstract content. And whilst pictures show an abundance of concrete details and rely heavily on the imagination to supply high-level content such a narrative structure, literary narratives present the reader with a selection of general claims and abstract content, relying on our imagination to fill it in with detail. For example, in the Genesis story the focus is on Abraham and his relationship with God; in Caravaggio, Isaac and the angel are made more prominent. Hence, different media offer different opportunities and constraints which artists can explore and play with, and that is part of the excitement of engaging with art in all the richness of its forms.
* * *
Recordings of the Fourth Meeting of the Northern Imagination Forum are now available on the NIF Youtube channel.