Behavioral and Neural Signatures of Visual Imagery Vividness Extremes: Aphantasia versus Hyperphantasia

Abstract

Although Galton recognized in the 1880s that some individuals lack visual imagery, this phenomenon was mostly neglected over the following century. We recently coined the terms "aphantasia" and "hyperphantasia" to describe visual imagery vividness extremes, unlocking a sustained surge of public interest. Aphantasia is associated with subjective impairment of face recognition and autobiographical memory. Here we report the first systematic, wide-ranging neuropsychological and brain imaging study of people with aphantasia (n = 24), hyperphantasia (n = 25), and midrange imagery vividness (n = 20). Despite equivalent performance on standard memory tests, marked group differences were measured in autobiographical memory and imagination, participants with hyperphantasia outperforming controls who outperformed participants with aphantasia. Face recognition difficulties and autistic spectrum traits were reported more commonly in aphantasia. The Revised NEO Personality Inventory highlighted reduced extraversion in the aphantasia group and increased openness in the hyperphantasia group. Resting state fMRI revealed stronger connectivity between prefrontal cortices and the visual network among hyperphantasic than aphantasic participants. In an active fMRI paradigm, there was greater anterior parietal activation among hyperphantasic and control than aphantasic participants when comparing visualization of famous faces and places with perception. These behavioral and neural signatures of visual imagery vividness extremes validate and illuminate this significant but neglected dimension of individual difference.

Keywords: aphantasia; autobiographical; hyperphantasia; imagery; neuroimaging.

Figure 1

Mean composite score (maximum = 54) for each group in the atemporal and future imagination tasks. The thick line within the violin plots display the median score, the thinner lines reflect the 25th and 75th percentiles. The top and bottom of the violin plots illustrate the lowest and highest scores in the sample. The plots were created using Prism GraphPad Prism v. 9 (https://www.graphpad.com/scientific-software/prism/).

Figure 2

Violin plots showing: (a) Mean number of internal details recalled for each time period; (b) mean rating (out of 21) for each time period; (c) mean number of external details recalled for each time period. The thick line within the violin plots displays the median score; the thinner lines reflect the 25th and 75th percentiles. The top and bottom of the violin plots illustrate the lowest and highest scores in the sample. The plots were created using Prism GraphPad Prism v. 9 (https://www.graphpad.com/scientific-software/prism/).

Figure 3

Seed to voxel connectivity analysis using the visual–occipital network as initial seed displaying areas of greater connectivity for hyperphantasic than aphantasic participants.

Figure 4

Areas more strongly activated in the hyperphantasic and control groups than the aphantasic group (left and right columns, respectively) in a subtraction of the perception from the visualization conditions.