Loss of imagery phenomenology with intact visuo-spatial task performance: A case of ‘blind imagination’

Abstract

The capacity for imagery, enabling us to visualise absent items and events, is a ubiquitous feature of our experience. This paper describes the case of a patient, MX, who abruptly lost the ability to generate visual images. He rated himself as experiencing almost no imagery on standard questionnaires, yet performed normally on standard tests of perception, visual imagery and visual memory. These unexpected findings were explored using functional MRI scanning (fMRI). Activation patterns while viewing famous faces were not significantly different between MX and controls, including expected activity in the fusiform gyrus. However, during attempted imagery, activation in MX's brain was significantly reduced in a network of posterior regions while activity in frontal regions was increased compared to controls. These findings are interpreted as suggesting that MX adopted a different cognitive strategy from controls when performing the imagery task. Evidence from experimental tasks thought to rely on mental imagery, such as the Brooks’ matrices and mental rotation, support this interpretation. Taken together, these results indicate that successful performance in visual imagery and visual memory tasks can be dissociated from the phenomenal experience of visual imagery.

Introduction

Asked to imagine a familiar item or recall a recent event, most people can call to mind an image that is less vivid than the original but has a visual ‘feel’. This expresses our capacity to experience visual mental imagery. There is a long-standing debate (see review in Bartolomeo, 2008) as to whether this mental experience plays a functional role in cognition: one view suggests that propositional knowledge is sufficient for supporting performance in imagery tasks (e.g. Pylyshyn, 2002), the other that imagery plays a key functional role (Kosslyn et al., 2006, Paivio, 1971). It could be that a range of cognitive strategies might be available to participants in these tasks. For example, imagery tasks might be performed using imagery or by using a propositional strategy. Both strategies might be available to healthy adults, although there is a wide variation in the extent to which people report the experience of visual imagery (e.g. Marks, 1973, McKelvie, 1995, Reisberg et al., 2003). Some subjects may, in other words, have a capacity for ‘blind imagination’, successful use of an alternative strategy to perform imagery tasks in the absence of the experience of imagery. This can be thought as analogous to the phenomenon of ‘blindsight’, characterized by successful performance in visuomotor tasks in the absence of visual perceptual experience.

We describe the case of an individual (MX) who reported the sudden loss of the experience of visual imagery in the absence of any other cognitive deficits. In previous published studies, the loss of visual imagery has been accompanied by impairments in performance on imagery tasks, for example in patients who show severe impairments on imagery tests but have intact perception (e.g. Farah, 1984, Bartolomeo, 2002, Kaski, 2002, Moro et al., 2008) and in patients who show pure representational neglect but no perceptual neglect (Guariglia et al., 1993, Beschin et al., 1997). In striking contrast, despite giving a persuasive description of his mental imagery deficit, MX was unimpaired on a range of standard tests of visual imagery. The main aim of this paper is to investigate the basis for the apparent dissociation between the lack of phenomenological experience in MX and his good performance.

Charcot and Bernard (1883, cited in Young & van de Wal, 1996) anecdotally described the case of an individual ‘Monsieur X’ who reported sudden onset of an inability to experience visual images of objects, such as monuments and buildings, and of familiar people such as close relatives. In this early case the visual imagery deficit was accompanied by some degree of prosopagnosia and clear visual memory impairments. Botez, Olivier, Vézina, Botez and Kaufman (1985) described the single case of a 38-year-old teacher presenting with an inability to experience visual images from birth in the absence of any overt deficits in perception, immediate or delayed visual memory. This example of a more selective deficit suggests that performance on visual imagery tasks might be normal in the absence of a phenomenal experience of visual imagery.

More comparable to MX are the two cases reported by Brain (1954) of sudden onset loss of visualisation, following head injury. One of these worked as a builder who, following a car accident, reported that he could no longer mentally visualise plans for new buildings, or even familiar routes, in stark contrast to his previous experience. Nevertheless he had no difficulty in drawing building plans. The second case sustained a head injury when tripping at work, and subsequently was unable to visualise recently viewed scenes or pages recently read from a book, although his reading and writing remained unimpaired. He also reported being unable to “see anything” when he closed his eyes. However, he could draw a plan of his house or a route from the station. These cases were not tested systematically on the imagery tasks developed from the 1970s onwards, nor, of course, was there any opportunity for detailed structural or functional neuroimaging. Nevertheless, as Brain himself recognised, his cases suggest the existence of dissociation between the (lost) phenomenological experience associated with visual imagery and performance on visual imagery tasks, which was largely intact.

Standard assessment of visual imagery abilities requires subjects to recall visual details, such as the relative length of animals’ tails, or to manipulate shapes ‘in the mind's eye’, for example to decide whether two shapes are in fact two views of a single object. Success in such tasks clearly requires access to a mental representation of the relevant material; it is not clear, however, that it always requires visualisation, the phenomenal experience of a visual image. There may be multiple routes to success in such tasks. For example, if asked about the colour of your mother's eyes you might answer by visualising her face and reading the answer off the resulting mental image, but it is possible to perform the same task by ‘just knowing’ the answer. These alternative strategies have been the focus of the polarised debate (Kosslyn et al., 2006, Pylyshyn, 2002) about whether mental imagery always serves a genuine cognitive function. We will return to the interpretation of our findings, and the imagery debate, in the discussion.