Neuroscience Research 201 (2024) 27–30

Contents lists available at ScienceDirect

Neuroscience Research
journal homepage: www.sciencedirect.com/journal/neuroscience-research

Short Communication

Revisiting the blind mind: Still no evidence for sensory visual imagery in
individuals with aphantasia
Rebecca Keogh a, b, *, Joel Pearson b
a
School of Psychological Sciences, Macquarie University, Sydney, Australia
b
School of Psychology, UNSW, Sydney, Australia

A R T I C L E I N F O A B S T R A C T

Keywords: The inability to visualise was given the name aphantasia in 2015 by Zeman and colleagues. In 2018 we published
Aphantasia research showing that fifteen individuals who self-identified as having aphantasia also demonstrated a lack of
Binocular Rivalry sensory visual imagery when undergoing the binocular rivalry imagery paradigm, suggesting more than just a
Visual Imagery
metacognitive difference. Here we update these findings with over fifty participants with aphantasia and show
that there is evidence for a lack of sensory imagery in aphantasia. How the binocular rivalry paradigm scores
relate to the vividness of visual imagery questionnaire (VVIQ) and how aphantasia can be confirmed is discussed.

In 2015 Zeman and colleagues published a paper about a group of 2022; Milton et al., 2021).
individuals who reported the inability to visualise, in other words, they The field of aphantasia is still in its infancy and the veracity of self-
had a ‘blind mind’s eye’ (Zeman et al., 2015), much like individuals reported aphantasia is still questioned by some due to the personal na­
described in Galton’s seminal work (Galton, 1880, 1883). Zeman et al. ture of visual imagery and heavy reliance on introspection and subjec­
(2015) gave this experience a name: congenital aphantasia. Since then tive reports (de Vito and Bartolomeo, 2016), although it is worth
much research and interest has burst forth about aphantasia, both in the pointing out that this argument can apply to any internal experience that
scientific arena, and from the general population (Bainbridge et al., relies on self-report. With the rapid increase of online studies and a
2021; Clemens, 2018; Dawes et al., 2020; de Vito and Bartolomeo, 2016; renewed interest in aphantasia, an important question keeps surfacing,
Grinnell, 2016; Jacobs et al., 2018; Keogh et al., 2021; Milton et al., how can we best identify or confirm aphantasia?
2021; Ross, 2016; Wicken et al., 2021; Zeman et al., 2016). In 2018 we published a study using the binocular rivalry imagery
Much of the research on aphantasia has centered around investi­ technique (Pearson, 2014; Pearson et al., 2008) showing that
gating whether having aphantasia impacts other cognitive abilities. self-identified individuals with aphantasia (N = 15) also lacked evidence
Thus far research has demonstrated that individuals with aphantasia can of visual imagery when measured using this paradigm (Keogh and
perform lab-based psychophysical and intelligence-type clinical visual Pearson, 2018), supporting their introspective reports. However, this
working memory tasks similarly and sometimes better than individuals was a relatively small sample, and we could not report on the likelihood
with imagery, potentially through using non-visual imagery strategies to that self-identified individuals with aphantasia would also show an
compensate (Keogh et al., 2021; Milton et al., 2021; Pounder et al., absence of sensory imagery using this technique.
2022). There is evidence that the ability to create attention templates Here we extend our previous findings by presenting the data we have
may be affected in individuals with aphantasia (Keogh and Pearson, collected at our laboratory over the years since this paper was published,
2021; Monzel et al., 2021), however, see also (Cabbai et al., 2023). The again using the binocular rivalry paradigm, in a larger group of self-
most convincing and potentially important cognitive function that may identified individuals with aphantasia (N = 55), including the 15 par­
be impacted by having aphantasia is autobiographical memory. ticipants from the 2018 study, plus participants who have participated
Research has demonstrated that individuals with aphantasia report in other lab-based published and unpublished studies (Kay et al., 2022;
reduced memories of the past (Dawes et al., 2020; Watkins, 2018; Zeman Keogh and Pearson, 2021; Keogh et al., 2021; Wicken et al., 2021). The
et al., 2020), as well as evidence of less vivid episodic memories of the binocular rivalry imagery task was the same as that reported in our
past when assessed using autobiographical interviews (Dawes et al., original 2018 paper (see Fig. 1). After undergoing an eye-dominance

* Corresponding author at: School of Psychological Sciences, Macquarie University, Sydney, Australia.
E-mail address: [email protected] (R. Keogh).

https://doi.org/10.1016/j.neures.2024.01.008
Received 30 June 2023; Accepted 23 January 2024
Available online 2 February 2024
0168-0102/© 2024 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
R. Keogh and J. Pearson Neuroscience Research 201 (2024) 27–30

general populations (N = 209, see pink violin plot, mean = 58.83, SD =

11.65), Mann-Whitney t-test used due to violation of normality and
differences in sample sizes (W = 3640.5, p < .001). We also ran a
Bayesian Mann-Whitney analysis and this provided strong evidence that
the aphantasia group had lower priming scores than the control group
(W = 3640.5, BF10 = 14.26). Next, we assessed whether individuals with
aphantasia, on average, showed any evidence of above-chance (50%)
priming. A one-sample t-test (Wilcoxon signed-rank test used for viola­
tion of normality) showed that, as a group, individuals with aphantasia
did not have priming that was significantly different from chance (V =
837, p = .284). This was further supported by a Bayesian analysis
Fig. 1. Binocular rivalry imagery paradigm timeline. Binocular rivalry was ach­ (Wilcoxon signed-rank analysis for violation of normality) which pro­
ieved using red-green anaglyph glasses. For ease of understanding the timeline vided evidence in favour of the null regarding whether individual’s with
shows what each eye is presented through the filters (e.g. left eye = green filter aphantasia binocular rivalry priming scores were greater than chance
and right eye = red-filter) however these images were physically shown in the (BF01 = 21.46). Finally, we used the new mean priming score we ob­
same location on the computer screen. During the imagery task participants tained from the 55 individuals with aphantaisa and compared this to the
were cued to imagine either a red horizontal or green vertical Gabor patch
bootstrap analysis run in the 2018 paper (see Fig. 2B, white bars). This
(letter R and G, respectively) on a black background for six seconds. Following
bootstrap analysis involved randomly sampling 15 participants from the
this they were presented with a binocular rivalry display comprised of a red-
horizontal and green-vertical Gabor patch and participants were asked to
general population and averaging their priming scores 1000 times. The
indicate which image was dominant (green = 1, mixed = 2 and red = 3). number of times the mean scores of these random samples were the same
or lower than the aphantasia mean priming score was then assessed (see
the black dashed line and white bars in Fig. 2B). We found that out of
calibration task participants completed the binocular rivalry imagery
1000 iterations, the new mean priming score (52.49%) was equal to or
paradigm. They were instructed to imagine either a red-horizontal or
less than 4 of the 1000 iterations, or p = .004. As this previous bootstrap
green-vertical grating (letter cues R and G respectively) for six seconds.
analysis used simulated samples of 15 to match our sample of those with
After this, they were presented with a brief (750 ms) binocular rivalry
aphantasia at the time, we wanted to update this to match the new
display and they were asked to indicate which image was dominant. The
sample size of 55 (see grey bars in Fig. 2B). When this was done 0 out of
percentage of trials in which the dominance of binocular rivalry was the
1000 iterations of a 55-person sample had a mean equal to or lower than
same as the cue prime (i.e. imagine red and see red) was measured
the aphantasia sample (the lowest bootstrapped mean in this case was
(known as percent primed). As with previous work, we interpret higher
55%). Taken together, these data replicate and extend our 2018 paper,
priming scores to indicate stronger sensory imagery. This is because a
supporting a lack of sensory imagery in self-identified aphantasia.
physically presented weak image also primes or facilitates subsequent
When our previous paper was published there was no consistent
binocular rivalry (Brascamp et al., 2007; Pearson et al., 2008), and
process used to assess aphantasia, other than self-identification so we
imagery shares a number of other priming characteristics with percep­
did not exclude any participants and thus we did not set any cut-off
tion (such as orientation (Keogh and Pearson, 2017; Pearson et al.,
criteria for our participants (this is also the case in Fig. 2A). With the
2008) and location specificity (Bergmann et al., 2016)), suggesting that
rise of interest in aphantasia the question of how best to confirm
visual imagery has sensory characteristics.
aphantasia has been raised frequently. To date the vividness of visual
In line with our previous findings, the priming scores (our measure of
imagery questionnaire (VVIQ (Marks, 1973) has been most frequently
imagery strength) in those with aphantasia (N = 55, see teal violin plot
used to ‘confirm aphantasia’ with two main approaches used in the
in Fig. 2A, mean = 52.49, SD = 8.51) were significantly lower than the

Fig. 2. Binocular rivalry priming data. A. Data shows binocular rivalry priming scores from the 2018 paper for aphantasia (left grey-scale violin plot) and general
population data (middle pink violin plot). The rightmost teal violin plot shows data from all individuals with aphantasia tested on the binocular rivalry paradigm to
date (including those from the 2018 paper). B. Frequency histograms (stacked) for the bootstrapping analysis’ from the general population data (Bins = 1) – white
bars represent the same bootstrapping as the 2018 paper for the 15 sample and the grey bars represent the new bootstrap analysis drawing 55 participants to match
the new aphantasia group sample size. For the analysis 15 or 55 subjects priming scores were randomly chosen, averaged, then returned to the main pool of general
population subjects. This was done 1000 times each for the 2 bootstrap analysis’. Data shows the distribution of the mean of N = 15 for 1000 iterations (white bars)
and mean of N = 55 for 1000 iterations (grey bars). The aphantasia groups mean from the 2018 paper is shown on the far left (grey dotted line), while the new data is
represented with a teal dotted line. C. Violin plots show binocular rivalry data scores for individuals with aphantasia with scores below 32 on the VVIQ (N = 51) split
into two groups: VVIQ = 16 (N = 32, left dark purple violin plot) and VVIQ = 17–32 (N = 19, right light purple violin plot). Each individual data point in the violin
plots represents an individual participant, and lines on violin plots represent medians and quartiles.

28
R. Keogh and J. Pearson Neuroscience Research 201 (2024) 27–30

papers that investigate aphantasia. One approach is to include anyone individuals with aphantasia had scores above 60% on the binocular ri­
who scores a 32 or lower on the VVIQ, which coincides with the highest valry priming paradigm, so it may be possible that these individuals
score given by self-identified individuals with aphantasia in the original have some form of ‘unconscious visual imagery’ that leads to priming or
Zeman paper (Zeman et al., 2015), equivalent to the imagery being perhaps the differences reside in the metacognitive process of
either vague and dim (rating of 2) or no image at all (rating of 1) on the self-evaluation of imagery vividness. It could be interesting for future
VVIQ. The second way researchers have labelled participants is by research to investigate whether these individuals who report having
separating them into those who have ‘aphantasia’ (those who score 16 aphantasia, but prime binocular rivalry above 60%, likewise might also
on VVIQ) and those with ‘weak or limited visual imagery’ (self-­ show evidence of visual imagery using other objective measures (e.g.
identified individuals with aphantasia who score between 17–32 on the pupillometry, skin conductance or even decoding in neuroimaging).
VVIQ) (Zeman et al., 2020). For this reason, we decided to use these two These data also suggest that dividing individuals with aphantasia
forms of classification to assess whether these different classifications into 2 sub-groups may not provide much further explanatory power
would be reflected in differences in binocular rivalry priming. when conducting analyses. This may be because individuals who have
Using the first criteria (scores less than 32 on the VVIQ), four par­ aphantasia but provide some 2 ratings on the VVIQ may be interpreting
ticipants in thissample were removed due to VVIQ scores greater than 32 ‘dim and vague’ to mean something completely different to someone
(33, 34, 37 and 39) and of these 4 participants 2 had priming scores with visual imagery. For example, some of the participants who attend
above 60% (mean = 57.08, SD = 13.09). The new group of individuals our laboratory and provide some ‘dim and vague’ ratings on the VVIQ
with aphantasia comprised of 51 participants with VVIQ scores of 32 or report that while they don’t see an image they may get a ‘feeling’ of the
below and had a mean binocular rivalry priming score of 52.07% (SD = spatial layout of a scene and thus a rating of 1 does not quite capture
8.09), which equates to a mean that was greater than or equal to 3 out of their experience and that is why they provide a 2. This lines up with
the 1000 iterations for the first bootstrap analysis or a p = .003 (see other work that finds that spatial imagery is not reduced in aphantasia
Fig. 2B white bars). (Bainbridge et al., 2021; Dawes et al., 2020). Future research should aim
Using the second criteria that has been previously used to separate to develop qualitative questionnaires that better reflect the phenome­
individuals with aphantasia, we split the data into two groups ‘self- nological experience of individuals with aphantasia in addition to more
identified individuals with aphantasia with a VVIQ score of 16′ and ‘self- behavioural and objective indexes of aphantasia.
identified individuals with aphantasia with VVIQ scores of 17–32′ (see
Fig. 2C dark and light purple plots respectively). The individuals with CRediT authorship contribution statement
aphantasia who scored 16 on the VVIQ (left dark purple violin plot in
Fig. 2C, N = 32, Mean = 51.32, SD = 7.57) did have a mean priming Keogh Rebecca: Data curation, Formal analysis, Investigation,
score that was numerically lower than the individuals who scored 17–32 Methodology, Project administration, Writing – original draft, Writing –
on the VVIQ (right light purple violin plot in Fig. 2C, N = 19, Mean = review & editing. Pearson Joel: Funding acquisition, Resources,
53.33, SD = 8.97) however, this difference was not significant (Mann- Writing – review & editing, Conceptualization, Methodology,
Whitney independent t-test W = 247.5, p = .14). This lack of an effect Supervision.
was also partially supported by a Bayesian analysis with anecdotal ev­
idence for a null effect BF01 = 1.65. Finally, priming in both groups was Acknowledgements
not significantly different to chance (16: V = 241, p = .44; 17–32: V =
121.00, p = .15) and a Bayesian analysis provided moderate evidence We would like to thank the participants with aphantasia who par­
for the null effect for both groups (16: BF01 = 10.93; 17–32: BF01 = ticiapted in this study for their time and valuable personal insights into
4.68). aphantasia.
Finally, due to an increase in online testing, many researchers have
asked the question, what proportion of people who report having no References
visual imagery would likewise also show a lack of sensory visual im­
agery? While all participants here attended the laboratory in person, Bainbridge, W.A., Pounder, Z., Eardley, A.F., Baker, C.I., 2021. Quantifying aphantasia
through drawing: Those without visual imagery show deficits in object but not
they all contacted us via email or signing up to our online database and spatial memory. Cortex 135, 159–172. https://doi.org/10.1016/j.
thus we think their results are likely representative of the participants cortex.2020.11.014.
who participate in online studies. For the participants who scored 32 or Bergmann, J., Genc, E., Kohler, A., Singer, W., Pearson, J., 2016. Smaller primary visual
cortex is associated with stronger, but less precise mental imagery. Cereb. Cortex 26
lower on the VVIQ, 12% of those (6/51) had a priming score greater (9), 3838–3850. https://doi.org/10.1093/cercor/bhv186.
than or equal to 60% (which is 10% above chance and also roughly the Brascamp, J.W., Knapen, T.H.J., Kanai, R., van Ee, R., van den Berg, A.V., 2007. Flash
average priming score found in many binocular rivalry studies suppression and flash facilitation in binocular rivalry. J. Vis. 7 (12) doi:Artn 12
10.1167/7.12.12.
measuring visual imagery). We also analysed the proportion of in­ Cabbai, G., Brown, C.R.H., Dance, C., Simner, J., Forster, S., 2023. Mental imagery and
dividuals who scored above 60% priming by separating them into those visual attentional templates: A dissociation. Cortex 169, 259–278. https://doi.org/
with a VVIQ score of 16 and those with VVIQ scores of 17–32. For those 10.1016/j.cortex.2023.09.014.
Clemens, A. (2018). When the Mind’s Eye Is Blind. Scientific American. Retrieved from
with a score of 16 on the VVIQ, 4 out of 32 (12.5%) had scores above
〈https://www.scientificamerican.com/article/when-the-minds-eye-is-blind1/〉.
60% (priming scores: 62.2%,67.65%, 67.5%,73.33% - mean: 67.67%), Dawes, A.J., Keogh, R., Andrillon, T., Pearson, J., 2020. A cognitive profile of multi-
while for participants with a VVIQ score of 17–32, 2 out of 19 (10.5%) sensory imagery, memory and dreaming in aphantasia. Sci. Rep. 10 (1), 10022
had scores above 60% (priming scores: 72.42%,79.21% - mean: https://doi.org/10.1038/s41598-020-65705-7.
Dawes, A.J., Keogh, R., Robuck, S., Pearson, J., 2022. Memories with a blind mind:
75.82%). To summarise, the proportion of people who report no imag­ Remembering the past and imagining the future with aphantasia. Cognition 227,
ery (VVIQ<33 +self-identification) and likewise show priming < 60% 105192. https://doi.org/10.1016/j.cognition.2022.105192.
equates to about 88%, while for VVIQ16 +self-identification equates to Galton, F., 1880. Statistics of Mental Imagery. Mind 5, 301–318.
Galton, F. (1883). Inquiries into human faculty and its development. Macmillan.
around 87% and VVIQ17–32 +self-identification equates to about 89%. Grinnell, D. (2016). My mind’s eye is blind – so what’s going on in my brain? New
However, it should be noted these two subgroups have smaller sample Scientist. Retrieved from 〈https://www.newscientist.com/article/2083706-my-min
sizes and hence should be revisited in time with larger datasets. ds-eye-is-blind-so-whats-going-on-in-my-brain/〉.
Jacobs, C., Schwarzkopf, D.S., Silvanto, J., 2018. Visual working memory performance in
In sum, we extended our previous findings (Keogh and Pearson, aphantasia. Cortex 105, 61–73. https://doi.org/10.1016/j.cortex.2017.10.014.
2018) to a larger group of individuals with aphantasia. Specifically, as a Kay, L., Keogh, R., Andrillon, T., Pearson, J., 2022. The pupillary light response as a
whole, this group of individuals did not show evidence of sensory visual physiological index of aphantasia, sensory and phenomenological imagery strength.
Elife 11. https://doi.org/10.7554/eLife.72484.
imagery when measured using the binocular rivalry paradigm. One
interesting finding however is that approximately 12% of self-identified

29
R. Keogh and J. Pearson Neuroscience Research 201 (2024) 27–30

Keogh, R., Pearson, J., 2017. The perceptual and phenomenal capacity of mental Pearson, J., Clifford, C.W., Tong, F., 2008. The functional impact of mental imagery on
imagery. Cognition 162, 124–132. https://doi.org/10.1016/j. conscious perception. Curr. Biol. 18 (13), 982–986. https://doi.org/10.1016/j.
cognition.2017.02.004. cub.2008.05.048.
Keogh, R., Pearson, J., 2018. The blind mind: No sensory visual imagery in aphantasia. Pounder, Z., Jacob, J., Evans, S., Loveday, C., Eardley, A.F., Silvanto, J., 2022. Only
Cortex 105, 53–60. https://doi.org/10.1016/j.cortex.2017.10.012. minimal differences between individuals with congenital aphantasia and those with
Keogh, R., Pearson, J., 2021. Attention driven phantom vision: measuring the sensory typical imagery on neuropsychological tasks that involve imagery. Cortex 148,
strength of attentional templates and their relation to visual mental imagery and 180–192. https://doi.org/10.1016/j.cortex.2021.12.010.
aphantasia. Philos. Trans. R. Soc. Lond. B Biol. Sci. 376 (1817), 20190688. https:// Ross, B. (2016). Aphantasia: How It Feels To Be Blind in your Mind. Retrieved from 〈htt
doi.org/10.1098/rstb.2019.0688. ps://www.facebook.com/notes/blake-ross/aphantasia-how-it-feels-to-be-blind-in-y
Keogh, R., Wicken, M., Pearson, J., 2021. Visual working memory in aphantasia: our-mind/10156834777480504/〉.
Retained accuracy and capacity with a different strategy. Cortex 143, 237–253. de Vito, S., Bartolomeo, P., 2016. Refusing to imagine? On the possibility of psychogenic
https://doi.org/10.1016/j.cortex.2021.07.012. aphantasia. A commentary on Zeman et al. (2015). Cortex 74, 334–335. https://doi.
Marks, D.F. (1973). Visual imagery differences in the recall of pictures. Br J Psychol, 64 org/10.1016/j.cortex.2015.06.013.
(1), 17–24. Retrieved from 〈http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? Watkins, N.W., 2018. A)phantasia and severely deficient autobiographical memory:
cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=4742442〉. Scientific and personal perspectives. Cortex 105, 41–52. https://doi.org/10.1016/j.
Milton, F., Fulford, J., Dance, C., Gaddum, J., Heuerman-Williamson, B., Jones, K., cortex.2017.10.010.
Zeman, A., 2021. Behavioral and neural signatures of visual imagery vividness Wicken, M., Keogh, R., Pearson, J., 2021. The critical role of mental imagery in human
extremes: aphantasia versus hyperphantasia. Cereb. Cortex Commun. 2 (2), tgab035 emotion: insights from fear-based imagery and aphantasia. Proc. Biol. Sci. 288
https://doi.org/10.1093/texcom/tgab035. (1946), 20210267. https://doi.org/10.1098/rspb.2021.0267.
Monzel, M., Keidel, K., Reuter, M., 2021. Imagine, and you will find - Lack of attentional Zeman, A., Dewar, M., Della Sala, S., 2015. Lives without imagery - Congenital
guidance through visual imagery in aphantasics. Atten. Percept. Psychophys. 83 (6), aphantasia. Cortex 73, 378–380. https://doi.org/10.1016/j.cortex.2015.05.019.
2486–2497. https://doi.org/10.3758/s13414-021-02307-z. Zeman, A., Dewar, M., Della Sala, S., 2016. Reflections on aphantasia. Cortex 74,
Pearson, J., 2014. New Directions in Mental-Imagery Research: The Binocular-Rivalry 336–337. https://doi.org/10.1016/j.cortex.2015.08.015.
Technique and Decoding fMRI Patterns. Curr. Dir. Psychol. Sci. 23 (3), 178–183. Zeman, A., Milton, F., Della Sala, S., Dewar, M., Frayling, T., Gaddum, J., Winlove, C.,
https://doi.org/10.1177/0963721414532287. 2020. Phantasia – the psychological significance of lifelong visual imagery vividness
extremes. Cortex.

30