Mental rotation of cubes with a snake face: The role of the human-body analogy revisited

ABSTRACT

Previous research has demonstrated repeatedly that the mental rotation of human-like objects can be performed more quickly than the mental rotation of abstract objects (a body analogy effect). According to existing accounts, the body analogy effect is mediated by projections of one’s own body axes onto objects (spatial embodiment), and the mental emulation of the observed body posture (motoric embodiment). To test whether motoric embodiment facilitates the mental rotation of human-like objects, we conducted an experiment using a snake-like object that had its own body axes but would be difficult to emulate. Twenty-four participants performed the mental rotation of snake-shaped cubes with or without a snake face as well as human-shaped cubes with or without a human face. Results showed that the presence of a face increased mental rotation speeds for both human-shaped and snake-shaped cubes, confirming both the human-body and snake analogy effects. More importantly, the snake analogy effect was equal to the human-body analogy effect. These findings contradict the motoric embodiment account and suggest that any object that can be regarded as a unit facilitates holistic mental rotation, which in turn leads to improved performance.

KEYWORDS:

• Mental rotation
• embodied cognition
• spatial transformation
• analogy

Acknowledgments

We thank Megumi Higashiyama, Keigo Nakahashi, Miho Nakanishi, Yukina Tanaka, Yujiro Toda, Haruho Umamichi, Takashi Yabuuchi, and Yuki Yamashita for collecting the data (their names are listed alphabetically). This work was supported by a JSPS Grant-in Aid for JSPS Fellows (Number 19J00072).

Disclosure statement

No potential conflict of interest was reported by the author(s).

ORCID

Hiroyuki Muto http://orcid.org/0000-0002-0007-6019

Notes

1 Taking a different approach, Jansen et al. (2012) reported that male soccer players showed shorter overall RTs for the mental rotation of human bodies than non-athletes, suggesting a potential role of motor expertise in this task. However, as they found no difference in mental rotation speeds (i.e., angle dependence) between groups, it seems premature to regard their results as evidence for motoric embodiment.

2 We selected these angular disparities to ensure a sufficient number of trials in each condition and excluded a 180° rotation for which the shortest rotation path was equivocal.

3 To guard against the violation of sphericity, we used Chi-Muller’s ε, by which the degrees of freedom and p-values were corrected. To keep the overall significance levels of α = .05 for multiple comparisons, we applied Holm’s sequentially rejective Bonferroni procedure.

4 We used g-prior distributions with r scale values of 1/2 and 1 for fixed and random effects, which are JASP’s default prior distributions.

5 Although this effect was significant, a BF value showed the opposite result (i.e., support for the null hypothesis). This inconsistency may be due to a multiplicity control method in JASP (see de Jong, 2019). Given the previous findings on the body analogy effect and the present results from the slope analyses, it should be rational to interpret this effect as existent.

Additional information

Funding

This work was supported by Japan Society for the Promotion of Science: [Grant Number 19J00072].