Visual imagery vividness declines across the lifespan; more pronounced in males

Visual imagery vividness declines across the lifespan. Erzsébet Gulyás et al. Cortex, July 9 2022. https://doi.org/10.1016/j.cortex.2022.06.011

Abstract: The capacity to elicit vivid visual mental images varies within an extensive range across individuals between hyper- and aphantasia. It is not clear, however, whether imagery vividness is constant across the lifespan or changes during development and later in life. Without enforcing the constraints of strict experimental procedures and representativity across the entire population, our purpose was to explore the self-reported level of imagery vividness and determine the relative proportions of aphantasic/hyperphantasic participants in different age groups. Relying on the frequently used Vividness of Visual Imagery Questionnaire, we collected data on a random sample of 2252 participants between the ages of 12 to 60 years. We found a novel developmental pattern that describes a declining ability to elicit vivid visual mental images in the group averages of different age groups from adolescence to middle age. This effect involves both a decreasing proportion of individuals with vivid visual imagery vividness and an increasing proportion of individuals with low imagery vividness as maturation (based on bone age assessments in adolescents) and ageing progress. These findings may shed some light on the developmental mechanisms of our internal, stimulus-independent processes, and might also help to determine genetic, maturational, and age-dependent factors in the cases of hyper- and aphantasia.

Keywords: imageryaphantasiadevelopmentmaturationlifespan

4. Discussion and conclusions

We examined whether visual imagery vividness is constant across the lifespan, or whether it changes during development and later in life. Our results show - for the first time - that visual imagery vividness declines with age, and this decline is more pronounced in males than females. Above average imagery vividness is common during the teenage years, while the proportion of hyperphantasics sharply declines from adolescence to middle age. Aphantasia, on the other hand, is non-existent in adolescents, and seems to become increasingly prevalent in the later years. Additionally, in 11 to13-year-old adolescents, advanced biological maturity (measured via bone age assessment) is correlated with weaker visual imagery vividness. We interpret these findings as evidence for the waning of imagery vividness as a function of chronological age between adolescence and middle age, and as a function of biological age in adolescents.

We believe that the discovered developmental changes in visual imagery vividness and in the prevalence of aphantasia are novel findings. In terms of visual mental imagery and ageing, there exist different experimental approaches, as well as studies on mental rotation, visuo-spatial internal representation, and visual working memory (for recent examples see: Craik & Dirkx, 1992; Dror & Kosslyn, 1994; Isaac & Marks, 1994; Wimmer et al., 2015) trying to define and model the phenomenon of visual internal representations. However, these studies approach mental imagery from a stimulus-response, functional perspective of short-term memory, while our study looks at the development of external stimulus independent visual mental imagery. The latter could avoid the potential biases of perceptual changes or decline due to ageing.

Our results contradict some of the already heterogeneous results of earlier studies measuring visual imagery vividness over the lifespan mentioned in the introduction (Isaac & Marks, 1994; Campos & Sueiro, 1993; Kemps & Newson, 2005; White et al., 1977; Wolmer et al., 1999). We noted that these studies use comparatively small sample sizes, limited age-based distribution, and other artefacts, which may contribute to inconsistent findings. The above described, age-related declining pattern of imagery vividness seems to provide a fresh and more coherent picture, and perhaps an alternative way of thinking about the background of internal representations. Our results may also shed light on the necessity of more representative surveys in this field to get more persuasive information about the maturational effects of the phenomenon.

Our observation on the remarkable distribution of the two extremes – namely hyper- and aphantasia – is also unique in the literature. Most previous studies on aphantasia look at its role in different cognitive functions, and potential impairments or compensative internal processes, mainly only in adult samples (for recent examples, see: Jacobs et al., 2018; Pounder et al., 2018; Keogh et al., 2021; Milton et al., 2021; Wicken et al., 2021). However, to understand the phenomenon more comprehensively, it would be essential to examine the nature of lifelong prevalence as well. The fact that both chronological and biological age are negatively correlated with visual imagery vividness seems to uncover a developmental process and indicates the existence of “developmental aphantasia” in addition to the potentially genetically based and acquired forms.

The other terminus of the visual imagery vividness spectrum is the phenomenon of hyperphantasia (Zeman, 2020). In contrast to aphantasia, this kind of extreme, ’offline-perceptual’ experience means to have abnormally strong, or photo-like, even eidetic imagery. According to the previous prevalence calculations, this mental representational ability is more frequent in the group of elementary-school-aged children, than among subjects in other age groups (Giray et al., 1976; Haber, 1979; Haber & Haber, 1964). Different theoretical approaches exist to explain this distributional pattern of which the most popular viewpoint is the developmental hypothesis (Haber, 1979; Haber & Haber, 1964). According to this assumption, extreme visual imagery vividness is an early capacity, modulated or lost by the progression of developmental processes during childhood. Nevertheless, this aspect could not necessarily provide a reliable explanation to any longitudinal observations, namely to the cases within the eidetic subjects that remained eidetically classified in the entire experimental time interval (Leask et al., 1969). Based on our findings, we would like to suggest that in addition to the genetically based neurodiversity along the visual imagery spectrum, a lifelong developmental pattern of decreasing visual imagery vividness is also part of the picture. The mechanism behind this decline and the neural background is not within the scope of our investigation here, however, we hope to facilitate a more detailed investigation of the neural correlates.

Inspired by our current findings we propose that the vividness of visual mental imagery is shaped by developmental factors, and there is a natural tendency for less vivid mental images with both maturation and ageing (see Fig. 2). Although we leave the possibility open that there might be a small proportion of individuals with genetically based hyper- or aphantasia whose imaging capacities are unaffected by maturational or developmental factors, we also claim the relevance of the developmental changes. As for future studies, this relevance might be twofold. On one hand, by acknowledging the changing distributions of imaging capacities, future studies might involve samples more representative for age and gender to determine the actual prevalence of either, potentially genetically based extremes. On the other hand, the clear declining tendency raises several questions about the developmental mechanisms that bring about such a remarkable change. For example, is there a difference between genetically based and developmental hyper- and aphantasia?

1. Download : Download high-res image (156KB)
2. Download : Download full-size image

Fig. 2

Instead of indulging in further exciting but unanswered questions, let us also note the limitations of our exploratory study that could be overcome in further investigations. First, despite the large number of participants in the adult age groups, our study cannot be considered representative, and it is not longitudinal. Therefore, we cannot completely rule out confounds related to random samples, and confounds that may include social, educational, technological, or lifestyle changes over time that may affect the spectrum of mental imagery vividness across the examined age groups. It would be very important to systematically study these potential confounds in future studies. Additionally, although unlikely, we cannot rule out the possibility that aphantasics from older age brackets were more inclined to participate in our study, e.g., by being overrepresented among the readers of the news portal where the invitation link of the study was published. Attrition bias might also be present if, e.g., mortality would decrease in the presence of aphantasia, or increase in the general and hyperphantasic population. The latter effect, that is, the relative decrease in the number of people with high imagery vividness in the population might be due to a confound with psychiatric or neurodegenerative diseases reducing life expectancy (Ji et al., 2019; Pearson et al., 2013; Pearson et al., 2015). It is also a shortcoming of this exploratory study that we used a self-report questionnaire that is subject to several response biases, and it may not be readily applicable in children.

Since it seems relevant to extend the current investigation to childhood, where over-reporting of imagery experience might be an issue, more objective measures involving lower levels of cognitive complexity are called for. For example, a no-report version (Frässle et al. 2014; Ziman et al., 2022) of the binocular rivalry dominance priming method (Milton et al., 2021; Pearson et al., 2008, 201; Keogh & Pearson 2018) might be a useful paradigm in forthcoming studies. The essence of this method is that there is an imagery instruction before each rivalry trial, and the induced mental image is expected to affect perceptual decisions about the multistable stimulus (Pearson et al., 2008). The facilitatory effect correlates with the subjective vividness reports (Pearson et al., 2011), and it is not present in subjects with aphantasia (Keogh & Pearson 2018). Although this method seems to deal with the subjectivity of self-reports, the rivalry paradigm is still short of objectivity since participants intentionally report their subjective percepts on each primed rivalry trial, involving criterion level problems, and individual results suggest on demand responses in a significant number of catch trials (Pearson et al., 2011). To solve this issue, a ‘no-report’ version of the rivalry paradigm, based on eye-tracking has been introduced (Frässle et al, 2014; Ziman et al., 2022). Another recently introduced method for the objective study of aphantasia uses the pupillary light response (Kay et al. 2022). During imagery of light or dark objects, the pupils react similarly as in natural vision: they dilate while imaging dark stimuli and constrict while imaging bright stimuli. Kay et al. (2022) found that the degree of imagery-evoked pupillary light response correlates with the proportion of successful priming in the binocular rivalry task and seems to be absent in aphantasia. Since intentional behavioural responses are not required (e.g., choices of questionnaire alternatives or button presses), while behavioural correlates of imagery vividness can be measured, these paradigms are good candidates for future objective studies of visual imagery vividness.

The lifelong changes in visual imagery vividness found in the current study should encourage future research to establish advanced, more objective techniques to measure vividness to determine exact age group standards that should help more precise prevalence estimations of hyper- or aphantasia. In addition to the prevalence estimations, the neural background of these conditions might be better revealed relying on the developmental information confirmed by objective methods, which should, in turn, help the understanding of visual imagery in general.

To sum up, we found a novel developmental pattern showing a declining ability to elicit vivid visual images as age increases from adolescence to middle age. This effect involves both a decreasing proportion of individuals with very vivid visual imagery and an increasing proportion of individuals with weak visual imagery as maturation and ageing progress. These findings may shed some light on the developmental mechanisms of our internal, stimulus-independent processes, and might also help to determine genetic, maturational, and age-dependent factors in the cases of hyper- and aphantasia.