Mental imagery and visual working memory abilities appear to be unrelated in childhood: evidence for individual differences in strategy use

Abstract

We addressed three research gaps related to Mental imagery (MI) in children. First, MI relies on depictive representations in adults, however evidence for a depictive theory of MI in children is lacking. Second, researchers have employed a four sub-component model (Image Generation, Image Maintenance, Mental Rotation, Image Scanning) to investigate the development of MI, however findings are mixed. Finally, shared mechanisms between MI and Visual Working Memory (VWM) are suggested in adult literature, yet this relationship has not been tested directly in children. Using a novel battery of MI tasks, we found evidence for depictive representations in both Experiment 1 (adults: N=58) and Experiment 2 (children age 6-11 years and adults: N=150) in that participants of all ages generated and maintained images of high vividness more often than low vividness. Moreover, we found participants make similar errors when estimating varying distances in both visual perception and mental imagery: participants of all ages underestimated distance, and this increased with increasing distance, thus providing further evidence to support a depictive theory in both children and adults. To address our second and third research questions, we report evidence to broadly support a separable-component model of MI and a dissociation between MI and VWM in both children and adults (Experiment 2). Our findings extend current understanding of development of MI from childhood to adulthood and broadly suggest the structure of MI in childhood mirrors that in adults. Moreover, the findings highlight the importance of considering individual differences in format of representations and strategy use when deciphering the relationship between MI and VWM in both children and adults.

Reimagining mathematics: The role of mental imagery in explaining mathematical calculation skills in childhood

“…Recent research has extended previous findings by adapting a battery of tasks to investigate individual differences in the visual precision of MI developmentally throughout the primary school years, i.e. from what age can children generate and manipulate images of high visual precision (Bates & Farran, 2020). In this study it was found that children from the youngest age tested (age 6 years) can generate and maintain highly visual mental images and these skills reach adult-like levels at approximately age 8 years in Image Generation and approximately age 10 years in Image Maintenance.…”

“…In this study it was found that children from the youngest age tested (age 6 years) can generate and maintain highly visual mental images and these skills reach adult-like levels at approximately age 8 years in Image Generation and approximately age 10 years in Image Maintenance. Moreover, the spatial accuracy of visual representations (Image Scanning) and how efficiently representations are transformed (Mental Rotation) were present from the youngest age tested (age 6 years) (Bates & Farran, 2020). Evidence has largely found that the sub-components of MI are not correlated in childhood (Bates & Farran, 2020;Kosslyn et al, 1990) supporting Kosslyn's separable-component model.…”

“…Moreover, the spatial accuracy of visual representations (Image Scanning) and how efficiently representations are transformed (Mental Rotation) were present from the youngest age tested (age 6 years) (Bates & Farran, 2020). Evidence has largely found that the sub-components of MI are not correlated in childhood (Bates & Farran, 2020;Kosslyn et al, 1990) supporting Kosslyn's separable-component model. However, it should be noted that in one study, researchers found positive associations between Mental Rotation and Image Scanning in age 6, 10 and adults but not age 4 and 8 years (Wimmer et al, 2017).…”

“…While one branch of MI research has focused on deciphering the format of representations, another branch has applied Kosslyn's component model of MI to investigate how MI develops throughout childhood (Bates & Farran, 2020;Kosslyn et al, 1990;Wimmer et al, 2015;Wimmer, Robinson, & Doherty, 2017 Kosslyn et al, 2006;Kosslyn et al, 1990). This theoretical model is supported by computational evidence suggesting that these sub-components are computationally distinct (Kosslyn et al, 1984), as well as neurological evidence that these sub-components are associated with activation in different brain regions (Kosslyn et al, 2004).…”

“…Degree of rotation differed across trials such that the target image below the line was presented at 0°, 45°, 90°, 135° or 180° (8 trials per degree of rotation; see Figure 2). The Image Scanning task used here was adapted from Wimmer et al (2016) by Bates and Farran (2020). Participants first viewed a map of an island containing a Lighthouse, Volcano, Hut, Pond and Tree for 45 seconds whereby they were instructed to memorise the island and the locations of the landmarks (see Figure 3).…”